Prekindergarten: Four Selected States Expanded Access by Relying 
on Schools and Existing Providers of Early Education and Care to 
Provide Services (09-SEP-04, GAO-04-852).			 
                                                                 
For nearly 40 years, the federal government has played a role in 
providing early childhood development programs for children of	 
low-income families through Head Start and other programs. Since 
1980, the number of states with preschool programs has also	 
significantly increased. While most of these programs have	 
targeted children at risk of school failure, more recently,	 
interest has grown in expanding these limited programs because of
the growing concern about children's readiness for school and	 
subsequent achievement. It has also been fueled by new research  
on early brain development that suggests the importance of early 
education and by the high rate of mothers in the workforce and	 
their need for early childhood services. In this context,	 
questions have arisen about how the various programs are	 
coordinated and what lessons have been learned from broad-based  
state preschool efforts. This work focused on four states that	 
have expanded their preschool programs to serve more children. In
these states, GAO addressed (1) how prekindergarten programs were
designed and funded, (2) the potential implications of these	 
program features for children's participation and other programs 
that serve four year-olds, and (3) the outcome data that have	 
been collected on participating children and families. To gather 
this information, GAO conducted site visits in four		 
states--Georgia, New Jersey, New York, and Oklahoma.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-04-852 					        
    ACCNO:   A12188						        
  TITLE:     Prekindergarten: Four Selected States Expanded Access by 
Relying on Schools and Existing Providers of Early Education and 
Care to Provide Services					 
     DATE:   09/09/2004 
  SUBJECT:   Child care programs				 
	     Data collection					 
	     Day care centers					 
	     Parents						 
	     Performance measures				 
	     Preschool education				 
	     Preschoolers					 
	     State programs					 
	     Program management 				 
	     Comparative analysis				 
	     Georgia						 
	     Head Start Program 				 
	     New Jersey 					 
	     New York						 
	     Oklahoma						 
	     HHS Child Care and Development Fund		 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-04-852

United States Government Accountability Office

GAO

                       Report to Congressional Requesters

September 2004

PREKINDERGARTEN

    Four Selected States Expanded Access by Relying on Schools and Existing
           Providers of Early Education and Care to Provide Services

GAO-04-852

Highlights of GAO-04-852, a report to congressional requesters

For nearly 40 years, the federal government has played a role in providing
early childhood development programs for children of low-income families
through Head Start and other programs. Since 1980, the number of states
with preschool programs has also significantly increased. While most of
these programs have targeted children at risk of school failure, more
recently, interest has grown in expanding these limited programs because
of the growing concern about children's readiness for school and
subsequent achievement. It has also been fueled by new research on early
brain development that suggests the importance of early education and by
the high rate of mothers in the workforce and their need for early
childhood services. In this context, questions have arisen about how the
various programs are coordinated and what lessons have been learned from
broadbased state preschool efforts.

This work focused on four states that have expanded their preschool
programs to serve more children. In these states, GAO addressed (1) how
prekindergarten programs were designed and funded, (2) the potential
implications of these program features for children's participation and
other programs that serve four year-olds, and (3) the outcome data that
have been collected on participating children and families. To gather this
information, GAO conducted site visits in four states-Georgia, New Jersey,
New York, and Oklahoma.

www.gao.gov/cgi-bin/getrpt?GAO-04-852.

To view the full product, including the scope and methodology, click on
the link above. For more information, contact Mamie Shaul at (202)
512-7215 or [email protected].

September 2004

PREKINDERGARTEN

Four Selected States Expanded Access by Relying on Schools and Existing
Providers of Early Education and Care to Provide Services

The expanded prekindergarten programs in Georgia, Oklahoma, New York, and
New Jersey had some similarities in their design features. For instance,
programs were offered at no direct cost to parents, regardless of family
income, and each state incorporated some level of collaboration with
community-based providers such as Head Start and large child care
facilities. Some key differences in their design features also existed.
For example, Georgia and Oklahoma had statewide programs providing
prekindergarten services to over half of their four-year olds, while New
York's and New Jersey's programs were more geographically targeted. States
and school districts also varied in offering full- or half-day
prekindergarten programs. States also varied in teacher qualifications,
the percentage of prekindergarten children served by community-based
providers, funding methods, and in the amount of funding per child.

State Per Child Funding for Prekindergarten Programs and Kindergarten
through Grade 12

New Jersey

New York

Georgia

Oklahoma

0 2,000 4,000 6,000 8,000 10,000 12,000

Half dayFull or half day Full or half day Full day Full dayFull day K - 12
K - 12 K - 12K - 12

Mixture of full-Half-dayKindergarten Full-day pre-K and half-day
pre-Kthrough 12

Sources: Data on per child funding for kindergarten through grade 12 were
obtained from the U.S. Department of Education,

National Center for Education Statistics; data on per child funding for
state prekindergarten programs were obtained from the Georgia Department
of Early Care and Learning, and the state departments of education in New
Jersey, New York, and Oklahoma.

Some program features had potential implications for the participation of
children and for early childhood programs. For example, none of the four
states required providers to transport all children to and from
prekindergarten, and many children were enrolled in half-day programs,
which officials believed might have limited the participation of children
from low-income and working families. Collaborations between programs and
community-based organizations generally permitted rapid program expansion
and were viewed as beneficial to early childhood programs. Finally, we
found few data to determine the impact of state prekindergarten expansion
on the availability or prices of child care.

While some data were available on outcomes for children who participated
in prekindergarten programs, less was known about their impacts on
families. For example, a study in Oklahoma showed that children who
participated made significant gains on several school readiness measures
relative to a comparison group of unenrolled children. However, none of
the

four states had measured effects on families, such as parents' work
effort.

Contents

Letter

Results In Brief
Background
Four Selected States Varied In the Design and Funding of their

Expanded Prekindergarten Programs Some Program Features Had Potential
Implications for Children's Participation and Other Early Childhood
Programs Some Data Have Been Collected on Outcomes for Participating

Children Concluding Observations Agency Comments

                                       1

                                      3 4

10

22

28 31 32

Appendix I Scope and Methodology

Appendix II GAO Contacts and Staff Acknowledgments 36

GAO Contacts 36 Staff Acknowledgments 36

Related GAO Products

Tables

Table 1: State Prekindergarten Program Characteristics, Benchmarks, and
Number of State Programs Meeting the Benchmarks 8

Table 2: Number of School Districts Statewide, Number of School Districts
with State Prekindergarten Programs, and Percentage of Age-Eligible
Children Participating in Five State Prekindergarten Programs during the
2003-04 School Year 13

Table 3: Variation in Selected Features of Five State Prekindergarten
Programs 16

Table 4: Primary Method of Program Funding, Estimated Number of Children
Participating, and Estimated State Spending for Prekindergarten Programs
during the 2002-03 School Year 17

Figures

Figure 1: Enrollment of Four-Year-Olds in State-Sponsored Prekindergarten
Programs 7

Figure 2: Estimated Number of Age-Eligible Children, Prekindergarten
Enrollments, and Head Start Enrollments in the Four States 12

Figure 3: State Per Child Funding for Prekindergarten Programs and
Kindergarten through Grade 12 19

Abbreviations

CBO community-based organization
CCDF Child Care and Development Fund
ECPA early childhood program aid
HHS U.S. Department of Health and Human Services
LEA local education agencies
NIEER National Institute for Early Education Research
SEA state education agencies
TANF Temporary Assistance for Needy Families
TPK targeted prekindergarten
UPK universal prekindergarten

This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.

United States Government Accountability Office Washington, DC 20548

September 9, 2004

The Honorable Edward M. Kennedy

Ranking Minority Member

Committee on Health, Education, Labor, and Pensions United States Senate

                       The Honorable Christopher J. Dodd

                            Ranking Minority Member

                     Subcommittee on Children and Families

Committee on Health, Education, Labor, and Pensions United States Senate

The Honorable Jack Reed United States Senate

For nearly 40 years, the federal government has played a role in providing
early childhood development programs for children of low-income families
through Head Start and other programs. Prior to 1970, only a few states
had prekindergarten programs, and these were mostly available to a small
number of children and targeted specific populations such as disadvantaged
children, those at risk of school failure, or those with special needs.
More recently, there has been interest in expanding these limited
prekindergarten programs because of the growing concern about children's
readiness for school and subsequent achievement. This interest has also
been fueled by advances in early brain development research that suggest
the importance of early education and by the high rate of mothers in the
workforce and their need for early childhood services. During the 2001-02
school year, about 40 states enrolled approximately 700,000 of the
nation's 7.8 million three- and four-year-olds in prekindergarten
programs, and spent over $2.4 billion to finance these programs. A few of
these states have invested significant resources to expand enrollments of
prekindergarten programs beyond targeted populations. They have also used
federal funds such as Title I of the Elementary and Secondary Education
Act, as amended by the No Child Left Behind Act of 2001, and Individuals
with Disabilities Education Act to support prekindergarten programs, and
have implemented these programs in a variety of settings such as child
care centers and Head Start agencies, as well as public schools.

Given the level of investment and the variety of preschool programs
currently receiving federal and state support, there is concern about
whether these programs are well coordinated and that lessons from
implementing broadly based state programs are documented. Accordingly, you
asked that we review programs in selected states that expanded their
prekindergarten efforts. This report presents information on (1) how state
prekindergarten programs were designed and funded in these states, (2) the
potential implications of these program features for children's
participation and for other programs that serve four-year-olds, and (3)
the outcome data that have been collected on participating children and
families.

To determine how prekindergarten programs were designed, we reviewed
program information and interviewed state officials, local education
program officials, as well as numerous Head Start grantees and directors
of large child care facilities, during site visits to four states
(Georgia, New Jersey, New York, and Oklahoma). These states were selected,
in part, because they were among a small subset of states that had
expanded their prekindergarten programs and aimed to serve all
four-year-olds whose parents wanted them to attend, regardless of family
income. We also considered states that had well-established
prekindergarten programs. Because New Jersey had two major preschool
programs, including the court-ordered Abbott program that was implemented
in New Jersey's 30 highest-poverty school districts, we report on a total
of five prekindergarten programs in four states.

To identify potential implications of the design of state prekindergarten
programs, we interviewed (1) state and local education agency officials;
(2) providers of prekindergarten services at community-based
organizations; and (3) federal, state, and local Head Start program staff.
We also reviewed selected data on child care availability in these states
and the nation. We assessed the data for reliability and reasonableness
and found them to be sufficiently reliable for the purposes of this
report. To determine what outcome data have been collected on
participating children and families, we interviewed state and local
education agency officials in each of the four states we visited,
contacted researchers, and reviewed two studies-a study of a school
district in Oklahoma and one statewide study in Georgia. We assessed the
methodologies used in these studies and determined that the methodologies
were of sufficient quality for use in discussion of most study findings,
and we limited our discussion to these findings. (App. I contains a more
detailed discussion of our scope and methodology.) We conducted our work
between October 2003 and

August 2004 in accordance with generally accepted government auditing
standards.

  Results in Brief

The four states we visited had developed expanded prekindergarten programs
that varied in design features and funding. The four states' program
features were similar in that each offered services at no direct costs to
parents and emphasized preparing children for school and preventing
failure in early grades. They also afforded the option of collaboration
with community-based providers. We found significant differences in other
aspects of the programs, such as geographic coverage, requirements for
teachers, and extent of reliance on community-based providers of early
childhood education. For example, Georgia had a statewide program that was
implemented in all of its school districts, while New Jersey targeted its
two prekindergarten programs in highpoverty areas-covering about 24
percent of its school districts. Teacher qualifications also varied among
the four states. Requirements for lead teachers ranged from an associate's
degree to a bachelor's degree with certification in early childhood
education. In providing their prekindergarten programs, states and school
districts also differed in the degree and type of collaborations they
established with community-based providers such as Head Start and private
child care centers. In terms of funding, the four states relied primarily
on state resources but differed in funding methods and amounts per child.
For example, New York financed its program using general revenues, while
Georgia relied on proceeds from a state lottery. With regard to funding
levels, New Jersey had the highest funding per child for its Abbott
prekindergarten-over $9,500 compared with less than $4,200 for each of the
other programs during the 2002-03 school year. All four states generally
used relatively small amounts of federal funds, such as those from Title I
or the Individuals with Disabilities Education Act, to support their
prekindergarten programs.

Some design features of state prekindergarten programs had potential
implications for children and early childhood programs in the four states
we visited. For example, both local officials and providers told us that
the lack of transportation and half-day programs may have affected
participation in prekindergarten programs for children of low-income and
working families. None of the four states we visited required
prekindergarten providers to transport all participating children.
Regarding the implications of state prekindergarten programs for other
early childhood program providers, three of the five programs served most
prekindergarten children through community-based organizations such as
Head Start and child care centers. State and local officials and
communitybased providers told us that collaborations were beneficial to
their

programs, but they also identified some challenges. In terms of benefits,
some states used collaborations to quickly expand their prekindergarten
programs by using existing age-appropriate facilities. In addition,
collaborations increased access to school district resources for some
community-based organizations. Some Head Start grantees received state
funds to provide prekindergarten services, while other grantees told us
they had begun serving more three-year-olds to offset the loss of
four-yearolds who participated in state prekindergarten programs
elsewhere, which potentially increased the number of poor children with 2
years of preschool education. At the same time, challenges remained for
some states and school districts-including the lack of expertise in
arranging formal contracts and maintaining effective collaborations with
community-based providers. Little empirical data were available to
determine the degree to which these broad-based state prekindergarten
programs affected the availability of and prices for child care, and the
anecdotal evidence we collected was mixed.

Some data have been collected on outcomes for participating children, but
little is known about outcomes for their families. In the school districts
we visited, we found that assessments of children's development were
incorporated in the curriculum and used to provide information to parents.
We also found two studies that provided information about the educational
outcomes of state prekindergarten programs in Oklahoma and Georgia. A
study conducted by Georgetown University collected data on prekindergarten
children in the Tulsa, Oklahoma, school district and found that children
who participated in the Tulsa program had significantly higher scores on
several school readiness measures than similar children in the comparison
group, who did not participate in the Tulsa prekindergarten program. A
second study, conducted by Georgia State University on Georgia's
prekindergarten program, reported that children who participated in the
state program made significant gains on several school readiness measures,
including language and problem solving, as did children in Head Start and
those enrolled in private prekindergarten programs. None of the four
states we visited reported collecting information regarding the impact of
their programs on families with respect to issues such as workforce
participation.

Background 	The growth in state prekindergarten programs has occurred for
various reasons, but three frequently cited reasons are (1) evidence of
the importance of early childhood to later development, (2) the high rate
of labor force participation by mothers of young children, and (3)
increased concern over school readiness and subsequent achievement. Much
has

been discovered about children's ability to learn more at an earlier age
than previously believed. The early childhood years are commonly portrayed
as formative. Between the first day of life and the first day of
kindergarten, development proceeds at a pace exceeding that of any
subsequent stage of life. Children from birth to age five engage in making
sense of the world on many levels: language, human interactions, counting
and quantification, spatial reasoning, physical causality, problem
solving, and categorization.1

Since the 1960s, the percentage of women in the labor force has increased
dramatically. In 1960, about 36 percent of women participated in the labor
force, and by 2000 this figure had increased to 58 percent. Moreover, in
2003, about 69 percent of women with children aged three to five (but none
younger) were in the labor force. This high rate of participation of women
in the labor force has resulted in more children enrolled in preschool
programs of varying quality and pressure being placed on schools to
provide before-and after-school programs.

To improve educational achievement for all children and reduce failure in
lower grades, many states and school districts are placing a greater
emphasis on the school readiness of younger children. For nearly a quarter
century, many states have developed or expanded their investment in
prekindergarten programs to increase the likelihood of children's success
in school. Prior to 1970 only 7 states funded preschool programs, and by
1988, 28 states had programs and total spending was $190 million;2 such
programs generally targeted economically and educationally disadvantaged
children. While most state-sponsored prekindergarten programs continue to
serve such children, a few states are in the process of expanding their
programs to include all four-year-olds, regardless of family income.
According to the National Institute for Early Education Research (NIEER),
40 states (and Washington, D.C.) had some form of state-sponsored
prekindergarten program in the 2001-02 school year and

1National Research Council and Institute of Medicine. 2000. From Neurons
to Neighborhoods: The Science of Early Childhood Development, Committee on
Integrating the Science of Early Childhood Development. Jack P. Shonkoff
and Deborah A. Phillips, eds. Board on Children, Youth, and Families,
Commission on Behavioral and Social Sciences and Education. Washington,
D.C.: National Academy Press.

2Mitchell, Anne W. 2001. "Education for All Young Children: The Role of
States and the Federal Government in Promoting Prekindergarten and
Kindergarten." Working paper series, Early Childhood Policy Research. The
Foundation for Child Development. New York, New York.

enrolled over 700,000 children, mostly four-year-olds.3 While states spent
more than $2.4 billion for prekindergarten during the 2001-02 school year,
10 states accounted for over 80 percent of this amount.

Generally, prekindergarten programs aimed to serve four-year-olds, but
according to NIEER estimates, most states served less than one-fifth of
all their four-year-olds (see fig. 1). NIEER estimated that about 80
percent of children served by state prekindergarten programs were
four-year-olds.4 During the 2001-02 school year, only two states (Georgia
and Oklahoma) enrolled more than 50 percent of their four-year-olds in a
state-sponsored prekindergarten program. Ten states (Alaska, Idaho,
Indiana, Mississippi, Montana, New Hampshire, North Dakota, South Dakota,
Utah, and Wyoming) had not initiated prekindergarten programs.

3NIEER, a unit of Rutgers University, supports early childhood education
policy through research. In the recent NIEER report The State of
Preschool: 2003 State Preschool Yearbook, enrollment and spending data
were based on 38 states providing information.

4NIEER estimated enrollment by age categories when states did not provide
data separately for three- and four-year-olds.

Figure 1: Enrollment of Four-Year-Olds in State-Sponsored Prekindergarten
Programs

Source: NIEER, The State of Preschool, 2003 State Preschool Yearbook.

Note: NIEER estimated enrollments by age categories when states did not
provide data separately for three- and four-year-olds. NIEER did not
report enrollments of state prekindergarten programs in Florida,
Minnesota, and Rhode Island.

The NIEER study also reported on characteristics describing the quality of
states' prekindergarten programs. Table 1 provides information on NIEER's
findings related to certain program characteristics, benchmarks, and the
number of state programs meeting the benchmarks that NIEER associated with
quality prekindergarten programs.

Table 1: State Prekindergarten Program Characteristics, Benchmarks, and
Number of State Programs Meeting the Benchmarks

                                                                    Number of 
Program characteristicsa           Benchmarksa            state programs b 
          Class size              20 children or fewer       
                                 At least one family support 
                                    service including parent 
Family support services              conferences or adult 
                               education, referral to social 
                                   services for families, or 
                                        information relating 
                                     to nutrition.           
      Staff-child ratio              1:10 or better          
                               Bachelor's degree for lead    
    Teacher qualifications              teachers             

Comprehensive curriculum standards	Developed specifically for
prekindergarten and cover the domains of language/literacy, mathematics,
science, social/emotional skills, cognitive development, health and
physical development, and social studies.

Source: NIEER, The State of Preschool, 2003 State Preschool Yearbook.

a NIEER identified 10 program characteristics and developed benchmarks for
each of the characteristics. The other program characteristics were
teacher specialized training requirements, assistant teacher degree
requirements, teacher in-service requirements, screening/referral
requirements, and meal requirements.

b Based on NIEER's summary of quality standards for a total of 44
programs, including some states with more than one prekindergarten
program, and Washington, D.C.

Among the 10 largest state prekindergarten programs, most met the
benchmarks for class size, family support services, staff-child ratio, and
teacher qualifications, and they were equally divided with respect to
comprehensive curriculum standards.

State-sponsored prekindergarten programs are expanding alongside existing
programs for young children, including Head Start, Title I, and private
child care programs. Head Start is a targeted program that mostly serves
children from low-income families.5 Head Start, administered at the

5Created in 1965, Head Start is the largest federal program that supports
early childhood education and was designed to help break the cycle of
poverty by providing comprehensive educational, social, health,
nutritional, and psychological services to low-income children.

federal level by the Department of Health and Human Services (HHS), is
implemented in local communities through grantees. These grantees include
community action agencies, school systems, for-profit and nonprofit
organizations, other government agencies, and tribal governments or
associations. The Department of Health and Human Services reported that
Head Start served just over 900,000 children nationwide during the 2003
fiscal year, and most were aged three and four; Head Start was funded at
about $6.7 billion, or about $7,366 per child.

In addition to administering Head Start, the federal government also
provides some limited support for early education programs through Title

I. Administered by the Department of Education (Education), Title I is the
single largest federal investment for elementary and secondary education.6
Its primary purpose is to help local education agencies and schools
improve the teaching and learning of children who are failing, or are most
at risk of failing, to meet challenging academic standards. In support of
that goal, Education reported that Title I was funded at about $11.7
billion during the 2003 fiscal year. Nearly 15 million students were
supported by Title I funds, and of these, about 2 percent (an estimated
313,000) were enrolled in prekindergarten programs during the 1999-2000
school year.7

The Child Care and Development Fund (CCDF) is the principal federal
program that supports child care for low-income families. CCDF is
administered by HHS, and each state receives an annual allocation that is
used to subsidize the child care expenses of low-income families with
children generally under age 13. CCDF subsidies can be used to obtain
child care from various types of providers, including child care centers

6Education also funds two other smaller early childhood education-related
initiatives. The Even Start program was designed to improve family
literacy and educational opportunities. The program is generally targeted
to parents who are not enrolled in school and their children through age
seven, who lack a high school diploma or its equivalent; the basic skills
necessary to function in society; or are unable to speak, read, or write
English. In program year 2000, Even Start served about 31,600 families and
41,600 children; in 2002, Even Start was funded at $250 million. Early
Reading First is a competitive grant program to support local efforts to
enhance the early language, literacy, and prereading development of
preschool age children, particularly those from low-income families,
through strategies and professional development that are based on
scientifically based reading research. In fiscal year 2003, nearly $75
million was awarded to 30 grantees.

7GAO, Title I Preschool Education: More Children Served, but Gauging
Effect on School Readiness Difficult, GAO/HEHS-00-171 (Washington, D.C.:
September 20, 2000) provides the most recent survey data available we
identified.

and family homes.8 In fiscal year 2002, CCDF was appropriated nearly $5
billion, and HHS reported that about 1.8 million children received
subsidies in an average month. As a condition of receiving CCDF funds,
states must conduct biennial surveys of child care providers, which are
considered by states when establishing reimbursement rates for providers
serving subsidized children. In addition to information regarding the fees
charged by providers for child care services, such surveys may provide
states with information about the type of child care they provide,
qualifications of the staff, the age groups of the children they serve,
and where they are located.

The four states we visited varied in the design features and funding of
their prekindergarten programs. Programs shared similar features such as
voluntary enrollment of children at no direct cost to their parents, but
differed in others. In addition, all five state programs permitted
collaboration with community-based providers. States varied in features
such as the teacher requirements for their prekindergarten programs.
States and school districts also differed in the degree and type of
collaborations they established with community-based agencies. Finally,
while states relied primarily on state resources, they reported some
differences in funding mechanisms and per child funding levels.

  Four Selected States Varied In the Design and Funding of their Expanded
  Prekindergarten Programs

    Selected States Designed Expanded Prekindergarten Programs to Enroll
    Children Voluntarily at No Direct Cost to Parents, but Differed in
    Geographic Breadth and Other Key Features

In the four states visited-Georgia, New Jersey, New York, and Oklahoma-we
found some similarities in prekindergarten programs.9 Over the last few
years, all four states had expanded their state-sponsored prekindergarten
programs and, as reported by NIEER in February 2004, were among only nine
states and Washington, D.C., that provided prekindergarten services to
more than 20 percent of their four-year-olds. All four states'
prekindergarten programs were provided at no direct cost to
parents-regardless of family income-and were offered on a voluntary basis;
children's enrollment was not mandatory. In addition, each program
emphasized preparation for school and incorporated the delivery of

8Child care centers typically provide for 12 or more children in a
nonresidential facility; family homes generally provide child care for a
small group of children in a provider's home.

9New York has two prekindergarten programs: the targeted prekindergarten
(TPK) and the universal prekindergarten (UPK). During the 2002-03 school
year, the TPK program served about 5 percent of the state's
four-year-olds. This report focuses only on the state's UPK initiative
that served about 23 percent of the state's four-year-olds.

prekindergarten services by community-based organizations as well as
schools. None of the states required that all providers offer
transportation services, although some providers did, and one state
offered reimbursement for some children when this occurred.10 Figure 2
shows the estimated number of age-eligible children in the state, and the
number of age-eligible children participating in prekindergarten and Head
Start programs in the four states we visited.11

10As a result of a state supreme court decision, the Abbott school
districts were required to provide transportation to children, if needed.

11The numbers of children who participated in state-sponsored
prekindergarten and Head Start are not mutually exclusive, that is, an
unknown number of children participated in both programs. Participation
data for Head Start refer to funded enrollment levels.

Figure 2: Estimated Number of Age-Eligible Children, Prekindergarten
Enrollments, and Head Start Enrollments in the Four States

Number of age-eligible children

300,000

                                    250,000

250,000

200,000

150,000

100,000

50,000

0 Georgia New Jersey New York Oklahoma States

Estimated number of age-eligible children (2003-04 school year)

Approximate number of pre-kindergarten enrollments (2002-03 school year)

Head Start enrollments (2003 fiscal year)

Sources: Estimates on numbers of age-eligible children statewide and
prekindergarten enrollments obtained from the Georgia Department of Early
Care and Learning, the state Departments of Education in New Jersey, New
York, and Oklahoma; Head Start data from the U.S. Department of Health and
Human Services, Head Start Bureau, Washington, D.C., and refer to funded
enrollment; actual enrollment may differ.

Note: With the exception of New Jersey, this analysis is based on
four-year-olds. In New Jersey, three- and four-year-olds were eligible in
some districts. Also, New York's prekindergarten program enrollment does
not include children participating in the targeted prekindergarten
program.

The states we visited differed in the extent of geographic coverage and
participation in prekindergarten programs. Three of the four states-
Georgia, New York, and Oklahoma-aimed to provide prekindergarten programs
to all four-year-olds in the state whose parents wanted them to attend.
While none of these states provided prekindergarten to all fouryear-olds,
Oklahoma and Georgia had the most widespread programs. During the 2003-04
school year, Oklahoma provided prekindergarten in 509 of its 541 school
districts to about 63 percent of its four-year-olds; Georgia provided
prekindergarten in all of its 181 school districts and to about 55 percent
of its four-year-olds. New York initially implemented much of its
universal prekindergarten program in school districts located

in the five largest cities in the state-Buffalo, New York City, Rochester,
Syracuse, and Yonkers. During the 2003-04 school year, about 80 percent of
the participating children attended prekindergarten in one of these five
cities. Overall, New York's prekindergarten program was offered in 190 of
its 680 school districts. In New Jersey's Abbott program, the state was
court ordered to provide prekindergarten to all three- and four-year-olds
who resided in the state's 30 highest-poverty school districts. In
addition, 102 non-Abbott early childhood program aid (ECPA) school
districts in high-poverty areas received funds for prekindergarten
programs. Combined, these two programs provided prekindergarten in 132 of
539 (24 percent) school districts in New Jersey. Table 2 provides
information on the extent of geographic coverage and percentage of
age-eligible children participating in prekindergarten programs in the
four states we visited.

Table 2: Number of School Districts Statewide, Number of School Districts
with State Prekindergarten Programs, and Percentage of Age-Eligible
Children Participating in Five State Prekindergarten Programs during the
2003-04 School Year

                                          Number of school Percentage of age- 
                             Number of      districts with  eligible children 
                                                     state 
                            school         prekindergarten   participating in 
                           districts                       
                   State     statewide            programs   prekindergartena 
                 Georgia           181                 181 
              New Jersey                                   
                 (Abbott           539                  30 
              districts)                                   
              New Jersey                                   
             (non-Abbott           539                 102 
                   ECPA)                                   
               New Yorkb           680                 190 
                Oklahoma           541                 509 

Source: Information was obtained from the Georgia Department of Early Care
and Learning and the state departments of education in New Jersey, New
York, and Oklahoma.

aPercentage of age-eligible children participating in prekindergarten was
calculated by dividing the number of children in prekindergarten by the
number of children residing in eligible school districts. In Georgia and
Oklahoma, all school districts were eligible for the state prekindergarten
program. In New York, 224 of the state's 680 school districts were
eligible for UPK funds. In New Jersey, 30 school districts were eligible
for the Abbott program. In addition, 102 non-Abbott high-poverty school
districts in New Jersey were eligible for non-Abbott ECPA prekindergarten
programs.

bThis table does not include four-year-olds served in New York's targeted
prekindergarten program, which served about 5 percent of the state's
four-year-olds.

The state-sponsored prekindergarten programs also differed in some of
their key design features. For example, the length of the program, in
terms of hours per day, ranged from 2.5 hours to 6.5 hours per day among
the four states. Full-day prekindergarten was provided in Georgia and New
Jersey's Abbott prekindergarten programs.12 The other three
prekindergarten programs-New Jersey's non-Abbott ECPA, New York, and
Oklahoma-allowed school districts to determine whether to offer full-day
or half-day programs.

The states also varied in their requirements for lead teachers, and two of
the five state programs (New Jersey's Abbott and Oklahoma) required
teachers to be certified in early childhood education.13 In New Jersey's
non-Abbott ECPA program, prekindergarten teachers could also hold
certification in elementary education. Beginning with the 2004-05 school
year, New York's state program required that all prekindergarten teachers
be certified, but certification could be in an area other than early
childhood education. In Georgia, lead teachers were required to hold at
least a technical diploma or degree, associate's degree, or Montessori
diploma. However, most lead teachers had at least a four-year-college
degree. As of May 2004, the Georgia Department of Early Care and Learning
reported that about 58 percent of its prekindergarten teachers were
certified in early childhood or elementary education and 21 percent

12The Georgia Department of Early Care and Learning also offered a
resource coordination grant program that was available to eligible
prekindergarten providers on a competitive basis so that they could
provide supportive services to children and their families in addition to
educational services. The key focus of the resource coordinator was to
involve parents in their child's educational development process by, among
other things, helping parents obtain needed health services for their
child and attending informational and child development seminars. Resource
coordinators may also help parents obtain community resources such as
information regarding the General Educational Development (GED) test,
employment counseling, and access to literacy classes. These grants were
made available to grantees to provide services to at-risk (category one)
children and families. Category one is defined as those children who
participated in one of the following programs: food stamps, Supplemental
Security Income (SSI), Medicaid, Temporary Assistance for Needy Families
(TANF), PeachCare for Kids, as well as those that were eligible for free
and reduced-price lunch. For the 2003-04 school year, over $18 million in
resource coordination grants was awarded, as of January 8, 2004.

13Certification in New Jersey's Abbott program, for example, requires that
teachers possess a preschool through grade 3 (P-3) endorsement. To qualify
for a P-3 endorsement, candidates must possess, at a minimum, a bachelor's
degree, a 2.5 grade point average, pass a state test when identified, and
complete a state-approved preparation program at one of the state's
colleges or universities. As of June 2004, New Jersey officials estimated
that about 93 percent of the teachers in the Abbott school districts have
met the certification requirement.

held four-year education-related or other degrees with some additional
training in early childhood education or development. Combined, about 79
percent of the lead teachers in Georgia had at least a four-year
educationrelated college degree.14

States and school districts established collaborations with communitybased
organizations differently and often relied on them extensively to provide
prekindergarten services to children. For example, Georgia had a
centralized program and the state's Department of Early Care and Learning
was directly responsible for establishing collaborations with
community-based organizations such as child care centers and U.S. military
bases. In contrast, in the other three states we visited, local school
districts had responsibility for establishing collaborations. In New York,
the state required that school districts use at least 10 percent of their
universal prekindergarten grant funds to serve children in communitybased
organizations, but statewide over 60 percent of four-year-olds were
participating in community-based prekindergarten programs during the
2002-03 school year. The extent of collaborations varied between the two
prekindergarten programs in New Jersey during the 2003-04 school year. In
the Abbott school districts, the state was ordered by the New Jersey
Supreme Court to provide full-day prekindergarten for all three-and
fouryear-olds; over 70 percent of these children were served by
communitybased providers.15 In contrast, in New Jersey's non-Abbott ECPA
school districts, only about 11 percent of the children received
prekindergarten from community-based providers. While state officials in
Oklahoma were supportive of collaborations, local school district
officials determined the role of community-based providers in their
prekindergarten programs. In Oklahoma, most children were enrolled in
prekindergarten programs in

14Of the remaining lead teachers in Georgia, about 21 percent held
technical or other degrees.

15In May 1998 the New Jersey Supreme Court mandated that children aged
three and four living in the 30 highest-poverty school districts in the
state receive a high-quality preschool education. The New Jersey
Department of Education funds a 6-hour, 180-day prekindergarten program,
and the New Jersey Department of Human Services funds a wraparound program
that provides daily before-and after-care and a summer program. In total,
the full-day, full-year program is available to eligible children for 10
hours per day, 245 days per year. The New Jersey Department of Education
reported that during the 200203 school year, the Abbott program enrolled
nearly 36,500 children at a cost of about $347 million for
prekindergarten. An official of the New Jersey Department of Human
Services also reported that about $103 million was spent for the
wraparound services during the 2003 fiscal year. In addition, school
districts can opt to have a 3-hour-per-day program for 20 days over the
summer break. This program would also be funded by the state Department of
Education.

public school buildings; the state did not know how many local school
districts collaborated with community-based organizations or the number of
children participating in them. These and other key differences in the
design and implementation of state prekindergarten programs are identified
in table 3.

 Table 3: Variation in Selected Features of Five State Prekindergarten Programs

New Jersey (Abbott)

New Jersey
(non-Abbott ECPA) New York Oklahoma

                            Program feature Georgia

Role of school School districts and School districts must School districts
                              Participating school

Participating school districts have option to contract with CBOs districts
and community-based organizations (CBOs) districts must use 10% of UPK
grant funds to serve children in CBOs

                 CBOs enter into   provide must provide some      
                contract with the   prekindergarten and level of  
               Georgia Department        contract with CBOs       
                                          prekindergarten         
                of Early Care and  wherever feasible services and 
                                                have              
               Learning to provide using a state model option of  
                 prekindergarten       contract establishing      
                                              contracts with CBOs 

     Maximum       20         15        No required        20         20      
class size                                                     
                                          maximum                 
    Hours per  6.5 hours   6 hours    Minimum of 2.75  Minimum of Minimum of  
       day                                                2.5         2.5     
                                     hours or 6 hours  hours or 5 hours or 6  
                                     at                 hours at  hours at    
                                       school district school          school 
                                                option district      district 
                                                       option          option 
Percentage      0          0             88             83         56      
       of                                                         
children in                                                    
    half-day                                                      
    programs                                                      
Curriculum  Sites used Determined Determined        Determined Determined  
                 state-   locally    locally            locally     locally   
               approved                                            based on   
               curricula                                             state    
                                                                  guidelines  

Content standards Statewide standards Statewide standards Statewide
standards Statewide standards 	Determined locally based on state
guidelines

Lead teacher Early childhood-Certification in early Certification in early
Teacher Certification in early

b de

qualifications 	related certification childhood education childhood or
certification childhood education or credentiala elementary

                                       c

                                   education

      Estimate for    57 percent 71 percent    11        63        Largely    
                                             percent   percent     school-    
     percentage of                                                     based; 
                                                                   percentage 
    prekindergarten                                                   in CBOs 
                                                                      unknown 
children served in                                           
          CBOs                                                  

Source: Information on program features was obtained from the Georgia
Department of Early Care and Learning and the state departments of
education in New Jersey, New York, and Oklahoma.

aGeorgia provides a variety of ways to meet the certification or
credentialing requirements, such as certification in early childhood or
elementary education; four-year-college degree in certain education fields
or four-year degree in psychology or social work with additional studies
in early childhood education or care; a technical institute diploma/degree
in early childhood education or care; associate degree in early childhood
education; or a Montessori diploma.

bBy September 2004, a bachelor's degree plus a state-approved preparation
program.

cBachelor's degree plus certification in early childhood or elementary
education.

dBeginning in the 2004-05 school year, a bachelor's degree plus must pass
two state teacher examinations (liberal arts and sciences test and
assessment of teaching skills).

e Bachelor's degree plus must pass general education, professional
teaching, and early childhood tests.

    State Funding Structure, Levels, and Stability Varied, and States Used
    Little Federal Funding

All four states relied primarily on state resources but differed in other
aspects of funding such as amounts per child, funding methods, and the
extent to which these methods and amounts provided for financially stable
programs. During the 2002-03 school year, enrollments and state spending
for prekindergarten services varied widely among the five state programs.
Based on data we collected from the states, spending ranged from
approximately $347 million for prekindergarten services in the 30 Abbott
school districts in New Jersey to about $30 million for the 102 non-Abbott
ECPA school districts in New Jersey. Table 4 provides information on the
primary methods of program funding, estimated number of participating
children, and estimated state spending among the five programs during the
2002-03 school year.

Table 4: Primary Method of Program Funding, Estimated Number of Children
Participating, and Estimated State Spending for Prekindergarten Programs
during the 2002-03 School Year

                                                              Estimated state 
                                               Estimated         spending for 
                                               number of      prekindergarten 
                         Primary method of      children programs (dollars in 
          State programs   program funding participating            millions) 

                 New Jersey -         General revenues:            
                       Abbott            school funding            
                                               formulaa   36,500         $347 
                      Georgia          Lottery proceeds   65,900         $252 
                     New York         General revenues:            
                                   annual appropriation   58,300       $195.4 
                     Oklahoma         General revenues:            
                                         school funding            
                                                formula   28,060        $66.4 
              New Jersey-non-         General revenues:            
                  Abbott ECPA            school funding            
                                                formula   7,200           $30 

Source: Information was obtained from the Georgia Department of Early Care
and Learning and the state departments of education in New Jersey, New
York, and Oklahoma.

aIn addition to receiving formula-driven funding, Abbott school districts
also receive preschool expansion aid. This aid is derived from the state's
general revenues and is provided to the Abbott districts to support the
required full-day program for three- and four-year-olds.

Per child expenditures for full-day and half-day prekindergarten varied
across the four states we visited and were consistently less than the
state's per pupil expenditures for kindergarten through grade 12. New
Jersey's Abbott districts had the highest funding per child for full-day
prekindergarten, relative to kindergarten through grade 12 funding.16 In
the remaining states, funding for full-day prekindergarten was much less
than the level of funding per child in kindergarten through grade 12. See
figure 3 for comparisons of per child expenditures for prekindergarten and
kindergarten through grade 12 in the four states we visited.

16While New Jersey's Abbott prekindergarten program per child funding was
more than double the other state programs, certain aspects of the program
contributed to the increased costs. For example, one feature (maximum
class size) increased program cost by roughly 25 percent by limiting
enrollment to 15 children. The New Jersey Department of Education also
initiated a number of activities to improve program quality. For example,
during the 2002-03 school year, the state provided a year-long training
program for 105 master teachers who, in turn, mentored and coached over
4,000 classroom teachers and assistant teachers in the Abbott school
districts. The New Jersey Department of Education also recommended that
the Abbott school districts employ one family worker for every 45
students. Family workers are to meet regularly with teachers and families
to identify specific concerns or needs, assist in locating resources in
the community, and support families in utilizing these resources.

Figure 3: State Per Child Funding for Prekindergarten Programs and
Kindergarten through Grade 12

$11,248

New Jersey

New York

Georgia

Oklahoma

0 2,000 4,000 6,000 8,000 10,000 12,000

Dollars

Kindergarten through grade 12

Full-day prekindergarten

Mixture of full- and half-day

Half-day prekindergarten

Sources: Data on per child funding for kindergarten through grade 12 were
obtained from the U.S. Department of Education,	National Center for
Education Statistics; data on per child funding for state prekindergarten
programs were obtained from the Georgia	Department of Early Care and
Learning, and the state departments of education in New Jersey, New York,
and Oklahoma.	

Note: State per child funding for prekindergarten programs was for the
2002-03 school year, and kindergarten through grade 12 expenditures were
for the 2000-01 school year.

Apart from New Jersey's Abbott and Georgia's prekindergarten programs, the
other state programs we examined were largely half-day. New Jersey's
non-Abbott ECPA program, New York, and Oklahoma permitted local school
districts to operate half-day prekindergarten. However, these states
differed in how they funded their half-and full-day programs. School
districts in New Jersey's non-Abbott ECPA program and New York received
the same amount per child whether they operated half-day or fullday
programs, and about 80 percent of the children attended half-day
prekindergarten in each of the two programs. In Oklahoma, local school
districts received about $1,743 per child (54 percent of the full-day
rate), and over half of the four-year-olds participating in the state's
prekindergarten program were enrolled in half-day programs during the
2002-03 school year.

The four states varied in how they funded their state-sponsored
prekindergarten programs, and officials in two states told us that the
financial outlook of their programs was stable. According to two New
Jersey state officials, because of the state supreme court decision and
subsequent court order, New Jersey was committed to providing a quality
prekindergarten program to all three-and four-year-olds who lived in the
Abbott school districts. In addition, funding for both New Jersey's Abbott
and non-Abbott ECPA prekindergarten programs was part of the school
funding formula. Oklahoma supported prekindergarten through the funding
formula, as it did for other school grades, and state officials told us
they believed that funding for the program was stable.

However, funding for prekindergarten in the other two states may be more
uncertain. For example, funding levels for New York's state-sponsored
prekindergarten had increased somewhat for the past 3 years but were
insufficient to allow the state to implement a universal prekindergarten
program available to all four-year-olds by the 2001-02 school year as
planned. During the same period, New York financed its program from
general revenue funds as a line item in the budget, and in 2003, the
program was targeted for elimination because of state fiscal shortfalls.
While avoiding elimination, limited increases in funding have restricted
the state's ability to expand the program over the past several years.
Most eligible districts participated in the program. However, about
two-thirds of the school districts were not eligible for the
state-sponsored prekindergarten program during the 2002-03 school year.

Georgia has historically relied on the state lottery to fund its
prekindergarten program. When the lottery was initially created, its
proceeds were set aside for three programs, including state-sponsored
prekindergarten. Currently, lottery funds are used to support
prekindergarten and a program to provide academic scholarships for
eligible high school graduates. However, over time, a greater percentage
of the lottery funds has been designated for the college scholarship
program than for prekindergarten. Additionally, lotteries recently began
in two neighboring states, and officials we interviewed were concerned
that Georgia's lottery proceeds may level off. State officials told us
that lottery funds may be insufficient to entirely support the
prekindergarten program by 2007, and the state has begun to look at stop
gap measures to protect lottery funding if needed in the future.

The four states reported using small amounts of federal funds to support
their prekindergarten programs; these amounts were generally small
relative to state funding levels. For example, two states-Georgia and

Oklahoma-used some prekindergarten monies to meet their CCDF matching or
maintenance-of-effort requirements.17 In fiscal year 2002, Georgia used
about $2.4 million in lottery funds for CCDF state matching and
maintenance-of-effort, which represented about 1 percent of the state
funding for prekindergarten. These funds were used for extended day
(before- and after-school) for Temporary Assistance for Needy Families
(TANF) eligible children participating in prekindergarten. Oklahoma used
about $2.1 million of its prekindergarten funds to meet CCDF
maintenance-of-effort requirements, which represented about 3 percent of
the state funding for prekindergarten. In fiscal year 2002, New York
transferred $61.3 million from the TANF program to the state
prekindergarten program, but this was done for only one year.18 None of
the other states used TANF funds to support the expansion of their
prekindergarten programs. While state officials told us that Title I,
Individuals with Disabilities Education Act, and Head Start program funds
were also used at the local level to support prekindergarten, they did not
know the exact amounts from these other federal sources.

17States receive CCDF funds from potentially four funding streams. Each
state's annual federal allocation consists of separate discretionary,
mandatory, and matching funds. A state does not have to obligate or spend
any state funds to receive the discretionary and mandatory funds. However,
to receive federal matching funds-and thus its full CCDF allocation-a
state must maintain its program spending at a specified level, referred to
as a state's maintenance of effort, and spend additional state funds above
that level.

18TANF, a block grant to states, provides temporary assistance to needy
families. In general, able-bodied TANF recipients, who receive cash
assistance, must participate in work or work-related activities after
receiving assistance for a maximum of 24 months, and there is a 5-year
lifetime limit on federal assistance. Beyond work, work-related activities
include education and training; job search; and participation in community
service. States may also use a portion of TANF funds for child care
services.

  Some Program Features Had Potential Implications for Children's Participation
  and Other Early Childhood Programs

Some prekindergarten design features had implications for children's
participation and early childhood programs in the four states we visited.
For example, both local officials and providers told us that
transportation and program hours may have affected access to
prekindergarten programs for children of low-income and working families.
State and local officials, along with community-based providers and Head
Start grantees, told us that collaborations were beneficial to their
programs and had allowed rapid expansion of state prekindergarten.
However, some challenges remained, such as the efforts needed to establish
and maintain effective collaborations. Finally, few empirical data were
available to quantify the effect of expanding state prekindergarten
programs on the availability and prices for child care, and the anecdotal
evidence we collected was mixed.

    Transportation and Program Hours May Have Implications for Participation

Program features, which varied across states and school districts, may
have affected participation, particularly for children of low-income and
working families. None of the four states required prekindergarten
providers to transport all participating children. Officials in some
school districts told us that the lack of transportation may have
decreased the participation of children from low-income and working
families, and 10 of the 12 school districts we visited did not provide
transportation to and from prekindergarten for all participating children.
Some school district officials primarily cited insufficient funding as a
reason for not providing transportation services. For example, in one
Oklahoma school district, children did not necessarily attend
prekindergarten classes at their neighborhood school; consequently, the
district would have incurred additional costs to transport children to
their designated school. One official in a rural school district we
visited in New York told us that more children from low-income and working
families would have participated in prekindergarten if transportation were
provided. However, costs prohibited the district from offering such
services. However, officials in the urban school district we visited in
New York did not view the lack of transportation as a barrier to
participation, as the prekindergarten programs were generally available in
proximity to children's homes or parents' jobs. In Georgia, the Department
of Early Care and Learning offered additional funding to providers who
opted to transport eligible children to and from the prekindergarten
program.19 In May 2004, the Georgia Department of Early Care and Learning
paid for the

19Transportation reimbursements were available for category one children.

transportation of 13,152 children. In New Jersey, the Abbott school
districts were required to provide transportation when needed.

In Oklahoma, where the majority of the participating children attended
half-day prekindergarten programs, school district officials told us that
the length of the school day affected participation. In all three school
districts we visited, local officials told us that shortened program hours
may have hindered the participation of children of low-income and working
families. Officials from one Oklahoma school district told us that the
combination of a half-day program coupled with the lack of transportation
to and from the prekindergarten program reduced the participation of
children from low-income and working families. In that district,
approximately 45 percent of the district's elementary school population
was eligible for free and reduced price lunch during the 2003-04 school
year, but only 29 percent of the children enrolled in prekindergarten were
eligible for free and reduced price lunch, indicating lower participation
of children from low-income families. Similarly, in another Oklahoma
school district, while about 84 percent of the district's student
population was eligible for free and reduced price lunch, 60 percent of
children who participated in prekindergarten were eligible for free and
reduced price lunch. However, officials in two Oklahoma school districts
told us that certain factors discouraged them from offering full-day
programs; for example, they were able to serve twice as many children with
half-day programs, rather than full-day programs, using the same resources
(classrooms and teachers). Additionally, one of these officials told us
that it would be difficult to implement full-day prekindergarten while the
school district only offered a half-day kindergarten program.

Finally, some school district officials told us that the location of the
program could also affect the participation of children of working
families. In particular, half-day programs without transportation could be
more appealing to low-income and working families if they were offered in
a child care center where the child could receive care for the duration of
the work day. One urban school district we visited primarily offered
half-day prekindergarten classes. However, the school district officials
told us that the prekindergarten classes were sometimes offered in
conjunction with other programs at the same location. A district official
and child care providers in this school district told us that this
arrangement met the needs of low-income and working families as children
would receive a full day of care. Officials and child care providers in
this school district told us that in order to offer a full day of care,
some child care providers supplemented the state-sponsored prekindergarten
program with other

monies, including Head Start funding, CCDF subsidies, and parent payments.

    State and Local Officials and Community-Based Providers Said Collaborations
    Were Beneficial to Their Programs, but They Also Identified Some Challenges

Although state and local officials, as well as staff of community-based
organizations, told us that collaborations were beneficial, some
challenges remained. Officials in all four states we visited told us that
such collaborations allowed them to serve more children, and three of the
five programs served most prekindergarten children in community-based
organizations such as Head Start and child care centers. In Georgia, New
Jersey, and New York, officials reported that they made extensive use of
collaborations because they wanted to implement the prekindergarten
programs quickly and schools were often at capacity. In New Jersey, the
state supreme court ordered the state to provide full-day kindergarten as
well as full-day preschool for three-and four-year-olds in the 30 Abbott
school districts. To implement the court order, the districts turned to
community-based organizations to accommodate the influx of children. In
New York, school districts were only required to use at least 10 percent
of their universal prekindergarten grant funds to serve children in
community-based organizations. Two school districts we visited in New York
served the majority of participating children in community-based
organizations (66 and 100 percent). However, in Oklahoma, where school
districts had been experiencing declining enrollment, collaborations with
community-based organizations were less prevalent, as districts were able
to accommodate prekindergarten children in public schools. Officials from
most of the school districts we visited that did use collaborations told
us that the collaborations allowed them to take advantage of the existing
early child care and education infrastructure, such as buildings,
equipment, and assistant teachers to increase program capacity and reduce
program costs.

Child care providers who partnered with state prekindergarten programs
generally had favorable experiences with collaborations as well.
Specifically, providers mentioned increased enrollment, improved quality
of programs, and increased access to school district resources as benefits
of their partnerships. Some providers who collaborated with the state
prekindergarten programs in Georgia, New Jersey, and New York told us that
they expanded their centers to serve more children of all ages-as some
parents enrolled the younger siblings of their prekindergarten children in
the same child care center. In addition, child care providers in all four
states told us that the overall quality of care had improved as a result
of collaborating with the state prekindergarten program. Some providers
attributed the improved quality to various factors, including a

greater focus on learning, the presence of credentialed teachers, and
higher standards for the children. Finally, some child care providers who
collaborated with state-sponsored prekindergarten programs told us that
the partnership allowed them increased access to school district resources
such as professional development and materials such as new computers for
the classrooms.

Head Start grantees also told us that collaborations were beneficial to
their programs. For example, in Georgia and New York, some Head Start
grantees who provided prekindergarten services stated they were better
able to serve children by leveraging state prekindergarten and Head Start
funds. These grantees told us they were able to expand program hours and
enrich the learning environment while still providing Head Start's
services, including establishing family partnerships.20 In Georgia, Head
Start grantees served 3,654 children, or just over 5 percent of the
children enrolled in the program during the 2004 fiscal year. In New York,
Head Start provided about 345 classrooms of prekindergarten-representing
about 9 percent of the total number of prekindergarten classrooms. Two
Head Start grantees told us that since the state prekindergarten program
served four-year-olds they began serving more three-year-olds. As a
result, children from low-income families could participate in 2 full
years of preschool.

While collaborations generally benefited early childhood programs, some
challenges existed in establishing and maintaining partnerships between
state-sponsored prekindergarten programs and community-based
organizations. State and school district officials told us that
establishing and maintaining collaborations took effort, required
expertise, and involved increased monitoring and technical assistance,
including financial guidance. For example, while one school district we
visited had a staff person responsible for establishing and maintaining
collaborations, another school district did not have such an expert and
was unsure about how to develop partnerships or arrange the formal
contracts needed to collaborate.

20The Head Start program performance standards on family partnerships
provide minimum requirements for grantees in areas such as family goal
setting; accessing community services and resources; parent involvement in
child development and education; parent involvement in health, nutrition,
and mental health education; and parent involvement in shared decision
making and community advocacy.

Child care providers also mentioned certain challenges such as
insufficient or uncertain funding. For example, in New York and Georgia,
where per child funding to community-based organizations had remained
fairly level for at least the past 3 years, some child care providers told
us that the per child funding was insufficient and they had to use other
funding sources to support the collaboration. In addition, we found that
challenges existed in establishing the collaborations with Head Start in
all four states. In both Oklahoma and New York, there was no formal
mechanism, such as a statewide contract to facilitate collaboration
between school districts and Head Start grantees, and the two programs
sometimes coexisted in the same community without the benefit of shared
resources. Two of the three school districts we visited in Oklahoma did
not collaborate with Head Start; the third served about 9 percent of its
prekindergarten children through collaborations with Head Start grantees.
Challenges in establishing collaborations with Head Start also remained in
New Jersey.21 In 2003, New Jersey's Department of Education and Department
of Human Services developed plans for including Head Start grantees as
partners in providing prekindergarten in the Abbott districts over the
following 3 years. However, many challenges remained to achieving this
goal including agreement on appropriate per child funding levels, as well
as challenges in aligning curricula and other coordination issues.

    Few Data Available to Determine the Impact of State Prekindergarten Programs
    on Availability and Prices of Child Care

While some community-based providers were initially apprehensive about the
potential impact of the widespread availability of states' prekindergarten
programs on the market for child care, we found few data to support this
concern. In the states we visited, neither state officials nor the child
care provider community had data regarding the effects of expanded
prekindergarten programs on the availability and prices of child care. The
available data were limited to child care market rate surveys, which were
conducted by states to obtain information needed to set reimbursement
rates for child care, and data collected every 5 years by the Census of
Service Industries on the number of tax filings by child day care

21In 2000, the New Jersey supreme court approved the use of community
providers such as Head Start as part of the Abbott preschool program
whenever practical. Abbott v. Burke, 748 A.2d 82 (N.J. 2000). Two years
later, the court discussed the problems encountered in coordinating the
two programs, including differing teacher qualifications, differing
program standards, and the costs associated with bringing Head Start up to
state standards. The court concluded that "districts should utilize Head
Start providers unless they are not able and willing to comply with Abbott
preschool standards or unless the cost of doing so is demonstrably more
expensive than other high-quality alternatives." Abbott Ex. El. Abbott v.
Burke, 790 A.2d 842 at 853 (N.J. 2002).

providers. Market rate surveys provided relatively recent data on prices,
but generally did not include sufficient data to isolate any effects of
prekindergarten and were not always collected in a comparable or reliable
form before and after prekindergarten expansion. In contrast, the state
level data currently available from the Census of Service Industries were
collected in a consistent fashion over time and across states, but were
available only for the period through 1997, just 2 years after significant
growth in the prekindergarten program in Georgia, the oldest expanded
program of the five we studied.22 The data indicate that the number of
small child care providers per 1,000 preschoolers in Georgia and the
nation as a whole followed similar growth trends from 1987 to 1997, the
years of available data. Further, in the same time period, the number of
child care centers per 1,000 preschoolers and the number of employees paid
by these centers increased in both Georgia and the nation (where
prekindergarten services were generally less available than in Georgia).
However, this does not prove that the expansion of prekindergarten
programs had no effect on the number of child care providers. For example,
perhaps the number of providers would have increased even more had
prekindergarten programs not been expanded.

The anecdotal evidence regarding the effects of prekindergarten programs
on the market for child care was mixed. For example, representatives of
the child care community mentioned some positive effects on the market for
child care, including the increased availability and accessibility of
highquality child care and early education for children from low-income
families. However, some child care providers in Georgia, New Jersey, and
Oklahoma told us that state programs had adverse effects on the business
of child care, but they were unable to provide us with supporting
documentation. According to child care providers, the care of three-and
four-year-olds was less costly than the care of infants and toddlers, and
the revenue generated from caring for the older children subsidized the
care of younger children and made up a significant portion of their
revenues. Child care providers also said that the enrollment of
four-year-olds in state prekindergarten programs could result in child
care centers raising prices to compensate for the loss of such revenues or
even going out of business. In addition, some child care providers in
Georgia told us that because state program funding did not cover the costs
of operating prekindergarten,

22Georgia's program was initially a targeted program. It expanded in the
early years (19921995) but remained targeted. In the 1995-96 school year,
the program became available to children regardless of family income. At
this time, the program went from enrolling 15,000 children to enrolling
44,000 children. The program has grown steadily since.

some centers had raised the rates for other services such as extended day
care.

However, any potential effects of prekindergarten on the price and
availability of child care may have been mitigated by certain design
aspects of the programs in the states we visited. For example, while the
majority of prekindergarten children in Oklahoma were served in public
school settings, the potential effects on the child care market may have
been mitigated because most children were in half-day prekindergarten
programs and some needed child care before and after the program. In
Oklahoma, the state Department of Human Services also provided a fullday
reimbursement for CCDF-eligible children who used child care for more than
4 hours a day. As a result, the state's half-day program appeared to have
minimal impact on child care providers. In New Jersey's Abbott districts
and Georgia, which had full-day prekindergarten, the classes were often
situated in community-based organizations. Consequently, many
four-year-olds who attended these programs remained in communitybased
settings, and child care providers maintained their enrollment of
four-year-olds. Furthermore, in the Abbott districts, the New Jersey
Department of Human Services provided additional funds to cover 4 hours of
child care, beyond the 6 hour educational program. As a result, some child
care centers were reimbursed for providing services for up to 10 hours per
day.

Some data have been collected on outcomes for participating children, but
little is known about outcomes for their families. In all the school
districts we visited, prekindergarten teachers routinely assessed children
and provided parents with information about their child's progress during
the school year. However, the states did not collect and analyze these
assessment data. We found two studies that provided information about the
educational outcomes of state prekindergarten programs on children in
Oklahoma and Georgia.23 One study focused on the Tulsa School District and
found that children who participated in the Tulsa prekindergarten program
had significantly higher scores on several school readiness measures than
children who did not participate in the program. A second

23William T. Gormley, Jr., and Ted Gayer, Promoting School Readiness in
Oklahoma: An Evaluation of Tulsa's Pre-k Program, Public Policy Institute,
Georgetown University, Washington, D.C., 2003; Gary T. Henry and others,
Report of the Findings from the Early Childhood Study: 2001-02, Andrew
Young School of Policy Studies, Georgia State University, August 2003.

  Some Data Have Been Collected on Outcomes for Participating Children

study analyzed statewide data regarding Georgia's prekindergarten program
and reported that children who participated in one of three programs
studied (Georgia's prekindergarten program, Head Start, and private
preschools) generally made significant gains on developmental skills
during the prekindergarten year. None of the four states we visited
reported collecting information regarding the impact of their programs on
families. State officials told us that prekindergarten programs increased
choices for families, but none reported knowing whether the
prekindergarten program had any effect on parents' work efforts.

    Teachers Regularly Assessed Children's Progress, although States Did Not
    Systematically Collect These Data

In general, teachers assessed the progress of the children's development
in the course of teaching prekindergarten using developmentally
appropriate assessments. We found that the types of assessments varied
across prekindergarten providers, and some providers used multiple types
of assessments. For example, assessments included checklists that rated
the child's progress on various developmental objectives, observational
records, as well as portfolio assessments, which consisted of a collection
of the child's work and projects that showed the child's progress
throughout the school year. In general, such assessments were used to
inform the teacher and provide information to parents during the school
year. None of the states required a particular assessment of children's
outcomes.24

State officials acknowledged the importance of collecting and analyzing
student outcome data. However, such analysis had not been systematically
conducted on a statewide basis in any of the four states we visited. The
outcome data that the teachers had were not necessarily in a form
conducive to collection and analysis by the states. Officials offered
various reasons for not collecting outcome data. In New York, officials
told us that there was no funding for large-scale data collection efforts
and they were awaiting the results of this year's fourth grade state test
to analyze the potential long-term effects of their prekindergarten
program, since some of these fourth graders had participated in the
prekindergarten program as four-year-olds. In New Jersey, state officials
told us that they planned to perform a program evaluation including
children's outcomes after the program had matured. In Oklahoma, state
officials told us that they did not

24Officials in Georgia reported that the Department of Early Care and
Learning was in the process of developing and piloting a standard
assessment process for all prekindergarten children.

collect outcome data for all children in the state, but limited
information regarding program outcomes was available in two school
districts. For example, we found that one school district we visited in
Oklahoma had collected and analyzed data on the outcomes of 22 children
over a 1-year period, 11 of whom participated in the district's
prekindergarten program.

    Two Studies Provided Some Information about the Outcomes for Children Who
    Participated in State Prekindergarten Programs

Two recent studies provided some information on outcomes for children in
two state prekindergarten programs. A study conducted by researchers at
Georgetown University analyzed the short-term effects of prekindergarten
on children in the Tulsa public schools and found positive effects of the
Tulsa program. In particular, the study found that children who
participated in the Tulsa prekindergarten program had higher scores on
both cognitive knowledge and language measures, and on measures of motor
skills, than did similar children who did not participate. Additionally,
the Tulsa study found that impacts tended to be larger for African
American and Hispanic children, and that there was little impact for white
children, although the authors discussed certain ceiling effects that may
have made it difficult to detect any impacts for white children as a
whole.25 The study also found that children who qualified for the full
free lunch program showed greater benefits than the population as a whole,
and benefits were larger for children from low-income families who
participated in full-day programs than those participating in half-day
programs.

The second study, sponsored in part by the Georgia Department of Early
Care and Learning and conducted by Georgia State University, also found
progress among children who participated in the Georgia prekindergarten
program during the 2001-02 school year. The study compared the progress of
children in three early education settings: private preschool, Georgia's
state prekindergarten program, and Head Start.26 Children participating in
the three programs performed at different levels upon entering the
programs. The study reported that at the beginning of prekindergarten,
children enrolled in Head Start demonstrated less mastery of certain
skills than did children in the Georgia prekindergarten program, who, in
turn,

25Ceiling effects can occur when children pretest at or near the highest
possible score on a given assessment. In such cases, if the same
assessment is used as a post-test, it is difficult to measure any gains
children have made.

26The study did not include a comparison group of children who did not
attend any preschool program.

Concluding Observations

scored lower than children in private preschools. The study found that
children in all three programs made significant gains over the course of 1
year, though in general, the gains made by the prekindergarten children
were not significantly different from the gains made by the other two
groups of children. By the end of prekindergarten, or at the beginning of
the kindergarten year, the relative rankings of the children from the
different programs had not changed. The researchers also matched children
with similar backgrounds to compare the effectiveness of Head Start and
Georgia's state prekindergarten program. When comparing the language,
communication, problem-solving, and basic mastery skill scores for the
matched samples of children, researchers noted one case-basic skill
mastery-in which the gap between scores of the state prekindergarten
children and the Head Start children widened to a statistically
significant level at the end of the prekindergarten year.

Citing benefits of prekindergarten, many states have made an investment in
the early education of young children, especially four-year-olds. Georgia,
New Jersey, New York, and Oklahoma have taken steps to expand early
educational opportunities for preschoolers. These four states offer
different approaches for consideration by other states that are
considering whether to expand the scope of their prekindergarten programs.
Given that states have limited resources, an opportunity exists to engage
community-based providers such as Head Start grantees and other early
education and care providers in the coordinated delivery of additional
prekindergarten services. Collaborations between school districts and
community-based organizations facilitate the coordination of child care
and early learning for preschoolers and can provide additional classroom
capacity. At the same time, these partnerships can help allay fears among
child care providers that prekindergarten programs would supplant the need
for community-based services.

It is also important to acknowledge the trade-offs of certain program
features. For example, while programs with limited hours may accommodate a
higher number of children within the same facility and may be less likely
to affect existing child care providers, they may create barriers to
participation for children of working families and show smaller effects in
school readiness of children. Also, prekindergarten programs may benefit
by collaborating with existing programs to maximize the efficient use of
limited state and federal education resources. However, such arrangements
may make it necessary for states and school districts to invest resources
to facilitate such coordination. Perhaps the biggest trade-off that states
face is whether the benefits of an expanded

prekindergarten program outweigh those of one that is more targeted.
Targeted programs have the advantage of giving more intensive services to
eligible children who may benefit most from prekindergarten, but such
programs exclude some children who might also benefit. Additional
information on outcomes for children in the most intensive programs,
particularly relative to children who do not receive comparable services,
may be helpful to other states considering varied types of prekindergarten
services, as would data on the benefits of half-day programs relative to
full-day programs.

The Departments of Health and Human Services and Education were provided a
draft of this report for review and comment. The Department of Health and
Human Services commented that given the discussions surrounding the "state
option" proposals being held during Head Start reauthorization, our report
is informative. Education's Executive Secretariat stated that the
department appreciated the opportunity to review the draft but was not
going to provide agency comments. Both agencies provided technical
comments, which we incorporated as appropriate.

  Agency Comments

We are sending copies of this report to the Secretary of Health and Human
Services, the Secretary of Education, relevant congressional committees,
and other interested parties. Please contact me on (202) 512-7215 if you
or your staff have any questions about this report. In addition, the
report will be available at no charge on GAO's Web site at
http://www.gao.gov. Other GAO contacts and staff acknowledgments are
listed in appendix III.

Marnie S. Shaul, Director Education, Workforce,

and Income Security Issues

                       Appendix I: Scope and Methodology

In conducting our review, we obtained and analyzed information from the
federal Departments of Education and Health and Human Services, state
agencies,1 and local education agencies (LEA). We visited four states, and
in each state we interviewed staff from state agencies and LEAs, and
generally included one urban, one suburban, and one rural school district
in each state-for a total of 12 school districts. We also interviewed
early childhood education and child care policy experts and reviewed
selected current literature on state-sponsored prekindergarten programs.

For our fieldwork, we considered states that (1) had expanded programs and
aimed to serve all children whose families wanted them to attend, (2)
served large numbers of children in their prekindergarten programs, and
(3) had well-established programs. We attempted to include varied program
models and gave some priority to states that had studied their efforts.

To determine how states designed their prekindergarten programs, we
interviewed state and local education officials and officials of 13
community-based organizations who were direct providers of state
prekindergarten services. We also reviewed documents related to states'
prekindergarten programs, including state laws, general program
information, state data on program participation and costs, curriculum
guides, content standards, and contracts governing collaborations with
community-based organizations. We obtained information to describe the
state-sponsored prekindergarten programs and reviewed the data for
reasonableness. We assessed reliability of specific information, including
estimates of age-eligible children and program expenditures, by
interviewing state officials in all four states about their data
reliability assessment processes. On the basis of this information, we
concluded that the data were sufficiently reliable for the purposes of
this report. In addition, we performed a detailed review of the
methodology of a National Institute for Early Education Research report
and found it to be sufficient for descriptive purposes. The data that were
used for background purposes were not independently verified.

To determine the potential implications of prekindergarten on other
programs that serve four-year-olds, we interviewed state child care

1In New Jersey, New York, and Oklahoma, we collected information and
interviewed officials from the state education agencies (SEA); in Georgia,
we collected information and interviewed officials from the Department of
Early Care and Learning, which administered the prekindergarten program.

Appendix I: Scope and Methodology

administrators, state and local Head Start association directors,
coordinators and program staff, as well as staff of local child care
centers. We also met with national and state representatives of the
National Child Care Association (an association of managers and owners of
child care centers) and related organizations, as well as state and local
staff of child care resource and referral offices. We also reviewed
selected national and state data on child care availability and prices.
Finally, we interviewed state officials regarding federal funds, including
Title I of the Elementary and Secondary Education Act, Temporary
Assistance for Needy Families, Individuals with Disabilities Education
Act, and the Child Care and Development Block Grant. We did not review the
expenses of providers of state prekindergarten services to ascertain the
extent, if any, to which federal funds or other revenues may have
subsidized provision of prekindergarten.

To determine what is known about the impacts of prekindergarten programs
on children and families in the states visited, we interviewed state and
local education officials and local policy experts. We also identified two
studies on children's outcomes that met our selection criteria: studies
that (1) analyzed student achievement and (2) compared prekindergarten
children with a control or comparison group of children who did not attend
the state-sponsored prekindergarten programs.2 To collect information
systematically, we developed a data collection instrument and examined
each study to assess the adequacy of the samples and measures employed,
the reasonableness and rigor of the statistical techniques used to analyze
them, and the validity of the results and conclusions that were drawn from
the analyses. A social scientist read and coded the documentation for each
study. A second social scientist reviewed each completed data collection
instrument and the relevant documentation to verify the accuracy of every
coded item. We found these two studies to be sufficiently reliable and
rigorous to include in our report.

2We identified a third study, the Rochester Early Childhood Assessment
Partnership 20022003 Annual Report, but did not review the study because
it did not include comparison groups and thus could not isolate the
potential effects of New York's universal prekindergarten program on the
children who attended.

Appendix I: Scope and Methodology

We did not identify any studies that assessed the impact of these
prekindergarten programs on working families such as workforce
participation.

We conducted our work between October 2003 and August 2004 in accordance
with generally accepted government auditing standards.

Appendix II: GAO Contacts and Staff Acknowledgments

GAO Contacts 	Betty Ward-Zukerman, (202) 512-2732, [email protected],
or Tim Hall, (202) 512-7192, [email protected]

Staff 	The following people also made important contributions to this
report: Nagla'a El-Hodiri, Shana Wallace, Alison Martin, Jean McSween,

Acknowledgments Barbara Hills, Susan Bernstein, Amy Buck, and Daniel
Schwimer.

Related GAO Products

Head Start: Better Data and Processes Needed to Monitor Underenrollment.
GAO-04-17. Washington, D.C.: December 4, 2003.

Child Care: States Exercise Flexibility in Setting Reimbursement Rates and
Providing Access for Low-Income Children. GAO-02-894. Washington, D.C.:
September 18, 2003.

Early Childhood Programs: The Use of Impact Evaluations to Assess Program
Effects. GAO-01-542. Washington, D.C.: April 16, 2001.

Title I Preschool Education: More Children Served, but Gauging Effect on
School Readiness Difficult. GAO-00-171. Washington, D.C.: September 20,
2000.

Early Education and Care: Overlap Indicates Need to Assess Crosscutting
Programs. GAO/HEHS-00-78. Washington, D.C.: April 28, 2000.

Preschool Education: Federal Investment for Low-Income Children
Significant but Effectiveness Unclear. GAO/T-HEHS-00-83. Washington, D.C.:
April 11, 2000.

Early Childhood Programs: Characteristics Affect the Availability of
School Readiness Information. GAO/HEHS-00-38. Washington, D.C.: February
28, 2000.

Education and Care: Early Childhood Programs and Services for Low-Income
Families. GAO/HEHS-00-11. Washington, D.C.: November 15, 1999.

  GAO's Mission

Obtaining Copies of GAO Reports and Testimony

The Government Accountability Office, the audit, evaluation, and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site (www.gao.gov). Each weekday, GAO posts newly
released reports, testimony, and correspondence on its Web site. To have
GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."

Order by Mail or Phone 	The first copy of each printed report is free.
Additional copies are $2 each. A check or money order should be made out
to the Superintendent of Documents. GAO also accepts VISA and Mastercard.
Orders for 100 or more copies mailed to a single address are discounted 25
percent. Orders should be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: 	Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

  To Report Fraud, Contact:

Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

Gloria Jarmon, Managing Director, [email protected](202)
512-4400Congressional U.S. Government Accountability Office, 441 G Street
NW, Room 7125 Relations Washington, D.C. 20548

Public Affairs 	Jeff Nelligan, Managing Director, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

                           PRINTED ON RECYCLED PAPER
*** End of document. ***