CD 824 B7abd1fc30bdf33cd824db3b102c4db0

User Manual: CD 824

Open the PDF directly: View PDF .
Page Count: 56

Download
Open PDF In Browser	View PDF

May 2006

Assessing Proposals for Preschool and
Kindergarten:

Essential Information for Parents, Taxpayers and Policymakers
By Darcy Olsen with Lisa Snell

POLICY
STUDY

344

Reason Foundation
Reason Foundation’s mission is to advance a free society by developing, applying, and promoting libertarian principles, including individual liberty, free markets, and the rule of law. We use journalism and
public policy research to influence the frameworks and actions of policymakers, journalists, and opinion leaders.
Reason Foundation’s nonpartisan public policy research promotes
choice, competition, and a dynamic market economy as the foundation for human dignity and progress. Reason produces rigorous, peerreviewed research and directly engages the policy process, seeking
strategies that emphasize cooperation, flexibility, local knowledge, and
results. Through practical and innovative approaches to complex problems, Reason seeks to change the way people think about issues, and
promote policies that allow and encourage individuals and voluntary
institutions to flourish.
Reason Foundation is a tax-exempt research and education organization as defined under IRS code 501(c)(3). Reason Foundation is supported by voluntary contributions from individuals, foundations, and
corporations. The views are those of the author, not necessarily those
of Reason Foundation or its trustees.
Copyright © 2006 Reason Foundation. Photos used in this publication
are copyright © 1996 Photodisc, Inc. All rights reserved.

Reason Foundation

Assessing Proposals for Preschool and Kindergarten:
Essential Information for Parents, Taxpayers and Policymakers

By Darcy Olsen with Lisa Snell
Executive Summary

roposals for universal preschool and all-day kindergarten are an increasingly popular policy
solution for everything from low academic achievement, to reducing crime, to lowering the
dropout rate. In summer 2005, a national task force co-chaired by Arizona Gov. Janet Napolitano
called for $8 billion annually in federal support for preschool.1 Similarly, in his 2006 response to
President Bush’s State of the Union Speech, Virginia Gov. Tim Kaine acknowledged universal
preschool as a silver bullet to help create a better future for the United States. Kaine said, “There’s
a Better way… Many states are working to make high quality Pre-Kindergarten accessible to every
family.”2
States are moving quickly to expand access to state-run preschool. According to Libby Doggett,
Pre-K Now’s executive director, states cumulatively have committed more than $14 billion to early
education.3 Arizona, New Mexico, Washington, South Carolina, Virginia, and West Virginia are all
considering various models of universal preschool, and Illinois Gov. Rod Blagojevich recently
announced plans to make Illinois the first state in the nation to offer universal preschool to both
three-and four-year-olds. California has a universal preschool initiative on the June 2006 ballot.
Nationwide, at least 40 states provide state funding for preschool programs, and at least 28
considered legislation to expand state-funded preschool programs in 2005.4 Three states—Georgia,
Oklahoma, and Florida—offer universal preschool.
The movement toward all-day kindergarten is also gaining popularity in the states. Currently nine
states mandate full-day kindergarten and seven states offer school districts financial incentives to
offer full-day kindergarten. Governors in Arizona, Indiana, and Massachusetts have made full-day
kindergarten a top legislative priority in 2006 and many other state legislatures are considering full-

day kindergarten proposals. For example, California is considering legislation that would make it
mandatory for all California school districts to offer full-day kindergarten by 2010.5 According to
the Education Commission of the States approximately 66 percent of kindergartners already attend
full-day kindergarten.6
California and Arizona are leading the charge toward universal preschool and full-day
kindergarten. California may become the national prototype for universal preschool. Hollywood
director Rob Reiner is promoting “Preschool for All,” a June 2006 ballot initiative, calling it “a
broad-based, multi-year, non-partisan advocacy campaign to achieve voluntary preschool for all
four-year-olds in California.”7
While universal preschool for all children sounds like a laudable goal, the Preschool for All Act
represents a de-facto institutionalization of preschool in California by creating a new, governmentmanaged $2.5 billion a year entitlement program that subsidizes the preschool choices of middleclass and wealthy families. Although it is a voluntary program, it would change the structure of the
current mixed-provider preschool market into a state-controlled monopoly.
California’s Preschool for All initiative would be financed by a 1.7 percent tax increase on
individuals who earn over $400,000 (or couples earning over $800,000), pushing the tax rate on
upper-income families to a national high of 12 percent.8 This new tax represents an 18 percent tax
increase on wealthy Californians.
Similarly, Arizona’s Governor Napolitano is representative of the national sentiment to incorporate
preschool education and full-day kindergarten into the current K-12 public school system. In 2004
Governor Napolitano released a School Readiness Action Plan that included the widely discussed
proposal for state-funded all-day kindergarten and a lesser-discussed plan for “state-supported
preschool.”9 Speaking before the National Task Force on Public Education, the governor said her
aim was “ensconcing early care and education as a lockstep component of public schooling.”10 She
considers the plan a “starting point” for the state’s role in the “development of Arizona’s youngest
children.”11
Universal preschool advocates like Rob Reiner and Governor Napolitano argue that early schooling
improves academic achievement and offers children long-term academic and economic benefits.
Yet this study finds the evidence supporting those claims to be unfounded, at best.
To help determine the efficacy of early education programs, we examine the results of some of the
programs considered to be early education models—including, Perry Preschool, Chicago Child
Parent Studies, Abecedarian, and Head Start—and find the research to be flawed and therefore of
questionable value. We also review information from the National Center for Education Statistics,
which reports no lasting reading, math, or science achievement differences between children who
attend half-day and full-day kindergarten. We also examine the results of the National Assessment
of Education Progress in Georgia and Oklahoma, where universal preschool has been fully

implemented without quantifiable benefit. We find the widespread adoption of preschool and fullday kindergarten is unlikely to improve student achievement.
America’s flexible approach to early education gives children a strong foundation. Skills
assessment at kindergarten entry and reports by kindergarten teachers show a large and increasing
majority of preschoolers are prepared for kindergarten. The effectiveness of the current system is
also evident in early test scores. At age 10, U.S. children have higher reading, math, and science
scores than their European peers who attend the government preschools cited by advocates as
models for the United States. To the degree that the state remains involved in financing early
education, we recommend measures for transparency, program assessment, and improved
flexibility through individual student funding.

Reason Foundation

Table of Contents
Introduction...............................................................................................................1
What Do We Know? Understanding the Research .................................................................. 2

Ready or Not? An Overview of America’s Preschoolers...............................................6
Full-day or Half-day? The Kindergarten Decision ........................................................8
What Impact Do Preschool and Kindergarten Have on Student Achievement? A Historical
Overview ............................................................................................................................. 12

How Do U.S. Children Perform? An International Examination.................................14
Examining the Economic Case for Universal Preschool..............................................19
A. Perry Preschool: Can $1 Today Yield $7 Tomorrow? ........................................................ 19
B. RAND and The Chicago Child-Parent Centers .................................................................. 21
C. The Carolina Abecedarian Project .................................................................................... 22
D. Head Start ....................................................................................................................... 24

Evaluating Universal Preschool in Georgia and Oklahoma ........................................27
A. Experience in Georgia: $1.5 billion on Universal Preschool Bears No Fruit ....................... 27
B. The NAEP and Universal Preschool .................................................................................. 28

Recommendations ...................................................................................................30
A. Increase Transparency...................................................................................................... 30
B. Streamline Existing Duplicative Early Childhood Programs ................................................ 31
C. Assess Impact................................................................................................................... 31
D. Transform Current Spending into Grants.......................................................................... 32

Conclusion...............................................................................................................35
About the Author .....................................................................................................37
Related Reason Foundation Studies..........................................................................38
Endnotes..................................................................................................................39

PRE-K EDUCATION

Part 1

Introduction

he United States’ move toward more government preschool and kindergarten programs is not
unprecedented. In France, Italy, and the United Kingdom, there is nearly universal enrollment
of three- and four-year-olds in center-based institutions.12 A few states across the country have
adopted similar systems. Georgia created the first statewide universal preschool program for fouryear-olds in 1993, and Oklahoma, New York, and West Virginia have moved in a similar direction.
In 2002, Florida voters adopted a constitutional amendment requiring the state to provide free
preschool for every four-year-old child.13
According to the National Childcare Information Center the United States currently spends more
than $20 billion annually on various day care and early education programs, including Head Start,
preschool, and various daycare programs.14 As policymakers consider early education proposals,
we have the opportunity to examine research on preschool and kindergarten, review experience and
findings from domestic programs, and look to international data.
We find strong evidence that the widespread adoption of preschool and full-day kindergarten is
unlikely to improve student achievement. For nearly 50 years, local, state, and federal governments
and diverse private sources have spent billions of dollars funding early education programs. Many
early interventions have had meaningful short-term effects on grade-level retention and special
education placement. However, the effects of early interventions routinely disappear after children
leave the programs.15 The phenomenon, known as “fade out,” is important because it means that
early schooling may be immaterial to a child’s later school performance, or that the current school
system as structured is unable to sustain those early gains.
For mainstream children, there is little evidence to support the contention that formal preschool and
kindergarten are necessary for school achievement or are more advantageous than learning in a
traditional setting—and there is some evidence that day care and preschool can be detrimental.
Experience supports those findings. From 1965 to the present day, the United States underwent a
sea change in formal early education. Preschool and kindergarten, which were rarely used in the
late 1960s, are now the norm. Despite increased enrollment in formal early education programs,
student achievement has shown little to no improvement. To the degree that international test data

Reason Foundation
are instructive, America’s decentralized early education system is outperforming the European
model and excels in equipping students for superior achievement in the elementary years.
Implicit in arguments in favor of universal preschool and full-day kindergarten is the
presumption that the state should take more responsibility for educating young children. A
large majority of “child advocates” envisions something similar, with almost 7 of 10 saying
government policy should move toward a universal, national system similar to those of many
European countries. Most parents feel otherwise. More than 70 percent of parents with young
children say it is their responsibility to pay the costs of caring for their children, and only one
in four would move toward a universal system paid for by the government. Also, a majority of
low-income parents (those earning no more than $25,000 per year) believes that bearing the
cost is their responsibility and not society’s. The public opinion research organization, Public
Agenda, reports, “At the most basic level, parents of young children believe that having a fulltime parental presence at home is what’s best for very young children, and it is what most
would prefer for their own family.”16
Universal preschool advocates often attempt to address parents’ concerns by saying participation in
the programs will be voluntary. Today, all 50 states have compulsory attendance laws, applying
generally to children between the ages of 5 and 18, and many policymakers have been forthright in
calling for extending compulsory education to preschoolers.
For example, in 2001, District of Columbia Councilman Kevin Chavous proposed the
“Compulsory School Attendance Amendment Act” to make school compulsory for every
preschool-aged child in the nation’s capital.17 The Honorable Zell Miller, former U.S. senator and
Georgia governor, has also expressed a preference for mandatory enrollment, saying, “If I had a
choice of pre-K or 12th grade being mandatory, I’d take pre-K in a second.”18 For many people
who are convinced that preschool is a necessity, mandatory attendance becomes the next logical
step. As one prominent Vermont legislator explained when he proposed a study on the cost of
compulsory preschool for three- and four-year-olds, compulsion is the only way to guarantee that
children have an equal opportunity for education.19
Fundamentally, the preschool and kindergarten debate is not about the effectiveness or expense of
the programs. At heart is the question of in whose hands the responsibility for young children
should rest. On that question, plans to entrench the state further into early education cannot be
squared with a free society that cherishes the primacy of the family over the state.

What Do We Know? Understanding the Research
Policymakers are interested in early education for several reasons. Some proponents see preschool
and kindergarten as a politically palatable way to subsidize day care.20 The primary argument made
by many policymakers, including Arizona Gov. Janet Napolitano, Arizona State Superintendent of
Public Instruction Tom Horne, and the Arizona State School Readiness Board, is that more early

PRE-K EDUCATION
learning will provide the experiences and environment necessary to promote the healthy
development of children, leading to subsequent school achievement:

State Superintendent of Public Instruction Tom Horne writes, “Studies show that a dollar spent
on academically oriented all-day kindergarten can equal more than $7 or $8 spent in later
grades in producing the same academic progress.”21

Gov. Janet Napolitano writes, “Extensive research shows that full-day kindergarten improves
students’ reading, writing and math skills, and it contributes to lower dropout rates.”22

The State School Readiness Board writes, “Full day kindergarten can lower grade retention,
improve language and math skills, lead to higher achievement test scores in eighth grade, and
improve attendance and social skills.”23

These sentiments are echoed by leaders in other states advocating for universal preschool.
California’s Superintendent of Public Instruction Jack O’Connell has argued that kids in highquality universal preschool have a strong foundation in literacy and reading. As he wrote in a
February 13th commentary in support of California’s universal preschool initiative, “perhaps most
important, these lucky kids have a good foundation for learning to read—the skill they will most
need to thrive through high school and beyond.”24
Unfortunately, most of the research informing those statements is limited in its applicability to
mainstream middle-class students and plagued by methodological shortcomings (including small
sample size, high attrition rates, infrequent random selection, and infrequent use of comparison
groups). Some of the research has been discredited.
For instance, Superintendent Horne suggests that one dollar invested in full-day kindergarten can
save seven dollars in later years. Although he does not specify, this figure appears to be based on a
flawed cost-benefit analysis from one study of 123 children conducted from 1962–1965, which
independent peer reviewers found to be compromised by significant sampling and methodological
errors. It also lacks the ability to inform the preschool discussion for mainstream children because
it included only children at risk of “retarded intellectual functioning.”25 Further undermining
confidence in the results is the fact that its findings have never been replicated.
Taken as a whole, a review of the research shows that many early interventions have had
meaningful short-term effects on disadvantaged students’ cognitive ability, grade-level retention,
and special-education placement. However, most research also indicates that the effects of early
interventions disappear after children leave the programs.26
This finding helps explain how two researchers can look at the same study and reach different
conclusions: The National Center for Education Statistic (NCES) studies, for instance, which have
received significant press coverage and are discussed later in detail, show a slight advantage for
full-day kindergartners over half-day kindergartners as measured at the end of the kindergarten
year. Critically, however, they show no differences in academic achievement between the two
groups by the end of third grade.

Reason Foundation
Recent research conducted by UC Santa Barbara has also confirmed the “fade out” phenomenon
for preschool children. A February 2006 study by researchers Russell W. Rumberger and Loan
Tran found no lasting academic impact from state-run preschool programs. They found that while
children enrolled in preschool had some moderate advantages in kindergarten performance, the
benefit dissipated by third grade.27
The phenomenon known as “fade out” is important to discussions of preschool and kindergarten
because it means that early schooling may not measurably affect a child’s later academic
performance. However, if fade out occurs, not because programs are ineffective, but because the
schools children later attend are unable to maintain those gains, then it is reasonable to conclude
that preschool and kindergarten will not result in lasting gains unless or until elementary and
secondary schools are significantly improved. Either conclusion points invariably to the need for
reform within the current school system.
As will be discussed later, in the few instances where research has shown the potential of early
intervention for improving children’s long-term outcomes, that research has been conducted on
severely disadvantaged children only in intense settings involving a level of intervention far
different from either preschool or kindergarten. For instance, in the widely cited Abecedarian
program, children were placed in the program as infants, at the average age of just over four
months old.
Importantly, most research has concentrated on children considered at risk of school failure, and
that research does not inform questions about the majority of mainstream students. The studies
conducted on mainstream children generally do not show benefits from early education programs.
According to David Weikart, past president of the High/Scope Educational Research Foundation
responsible for Perry Preschool, “For middle-class youngsters with a good economic basis, most
programs are not able to show much in the way of difference.”28 Similarly, a 2005 study on the
long-term economic benefits of high-quality preschool by the RAND Corporation found little
evidence to support the benefits of preschool for non-disadvantaged children. RAND
acknowledges that the one study analyzing mainstream children “found that children participating
in preschools not targeted to disadvantaged children were no better off in terms of high school or
college completion, earnings, or criminal justice system involvement than those not going to any
preschool.”29
A significant body of research shows that formal early education can be detrimental to mainstream
children. David Elkind, professor of child development at Tufts University and author of numerous
books on cognitive and social development in children and adolescents, explains,
The image of child competence introduced in the 1960s was intended to remedy some of the
social inequalities visited upon low-income children. But the publicity given the arguments of
child competence was read and heard by educators and middle-class parents as well…For this
reason it was uncritically appropriated for middle-class children by parents and educators.
While the image of childhood competence has served a useful function for low-income
children and children with special needs, it has become the rationale for the miseducation of
middle-class children…30

PRE-K EDUCATION
Elkind explains that children who receive academic instruction too early—generally before age six
or seven—are often put at risk for no apparent gain. By attempting to teach the wrong things at the
wrong time, early instruction can permanently damage a child’s self-esteem, reduce a child’s
natural eagerness to learn, and block a child’s natural gifts and talents. He concludes, “There is no
evidence that such early instruction has lasting benefits, and considerable evidence that it can do
lasting harm…If we do not wake up to the potential danger of these harmful practices, we may do
serious damage to a large segment of the next generation…”31
A November 2005 study by researchers at Stanford University and the University of California,
Berkeley analyzed data from more than 14,000 kindergartners from the National Center for
Education Statistics’ Early Childhood Longitudinal Study. They found evidence that preschool
hinders social development and created poor social behavior, such as bullying and aggression, and
a lack of motivation to take part in classroom activities. Those patterns for former center-based
preschoolers were the strongest among white children from high-income families and among lowincome black children. The study, "How much is too much? The Influence of Preschool Centers on
Children's Development Nationwide," found that children who attended preschool at least 15 hours
a week are more likely to display more negative social behaviors, such as acting up or having
trouble cooperating, than their peers. Children from better-off families were most likely to exhibit
social and emotional development problems, said UC Berkeley sociologist and co-author Bruce
Fuller.32
In a February 2006 study of Quebec’s universal preschool program examining more than 33,000
children between 1994 and 2002, economists from the C.D. Howe Institute find negative outcomes
for children enrolled in universal childcare. They write:
Several measures we looked at suggest that children were worse off in the years following the
introduction of the universal childcare program. We studied a wide range of measures of child
well-being from anxiety and hyperactivity to social and motor skills. For almost every
measure, we find that the increased use of childcare was associated with a decrease in their
well-being relative to other children. For example, reported fighting and aggressive behavior
33
increased substantially.
The notable absence of benefits for mainstream children coupled with evidence that early
education programs can be detrimental to their development should be of critical concern in light
of the fact that policymakers seek preschool and full-day kindergarten for all children, not just the
small percentage classified as being at risk for school failure.34

Reason Foundation

Part 1

Ready or Not? An Overview of
America’s Preschoolers

iscussions of preschool are premised partly on the notion that many children are inadequately
prepared for entry into kindergarten. For instance, the federal initiative Goals 2000
established “readiness” as the nation’s first education goal, stating, “By the year 2000, all children
in America will start school ready to learn.”35 Yet there is little agreement in child development
literature, among program proponents, or among parents about what children should know and
what skills they should possess or by what age, which makes defining “readiness” highly
subjective.36
Here we address the question of whether children are “ready” for kindergarten by examining: (1)
widely used proxy measures for assessing readiness; (2) concrete skills assessment at kindergarten
entry; and (3) how kindergartners perform on measures that kindergarten teachers say are the most
important for kindergarten preparedness. On these measures, data indicate that most children
entering kindergarten are equipped with the knowledge and traits required to begin the
kindergarten year.
In the Goals 2000 literature and elsewhere, researchers use preschool participation rates and the
frequency with which parents read to their children as two important indicators of readiness.37 By
those measures, a high and increasing percentage of American preschoolers are ready for
kindergarten. Data show only 5 percent of three-year-olds attended preschool in 1965—today, 42
percent attend. Sixteen percent of four-year-olds attended preschool in 1965—today, that figure is
68 percent.38
Data also show families engage their children in literacy activities regularly and with increasing
frequency. As measured from 1993 to 1999, the percentage of preschoolers who are read to three
or more times per week has increased from 78 percent to 81 percent. The percentage of
preschoolers who are taught letters, words, or numbers with equal frequency has increased from 58
percent to 64 percent. The upward trend is also present in the increasing percentage of preschoolers
who are taught songs or music, and have done arts and crafts with a family member.39

PRE-K EDUCATION
Therefore, according to the two common proxy measures of readiness—preschool enrollment rates
and early literacy activities—a majority and increasing number of preschoolers are prepared for
kindergarten entry. Although there may be room for improvement, the proxy data indicate that the
problem of under-preparedness is narrow and diminishing.
In 1998 the National Center for Education Statistics (NCES) began conducting the Early
Childhood Longitudinal Study (ECLS-K), which assessed 22,000 children at kindergarten entry
and most recently reported on those students through the third grade. The study is the only one of
its kind, using a nationally representative sample of children, and conducting a longitudinal and
multivariate analysis that is a requirement for assessing the long-term benefits of early education
and kindergarten programs. Researchers Nicholas Zill and Jerry West explain,
Until recently, we have lacked systematic information about what children know and can do at
school entry. The data that have been available depended on reports about children’s skills
from the parents of preschool children, rather than on direct assessments of the children
themselves. With the launching of the U.S. Department of Education’s Early Childhood
Longitudinal Study, Kindergarten Class of 1998–99 (ECLS-K) in the fall of 1998, however,
measures of the knowledge, skill, health, and behavior of a large and nationally representative
sample of American kindergartners are available.40
The NCES assessment allows researchers to move beyond proxies into specific, verifiable skills.
According to the first national assessment of the skills and traits children possess as they enter
kindergarten—“America’s Kindergartners”—U.S. kindergartners have a strong start. In terms of
concrete literacy development, 82 percent of children entering kindergarten have basic familiarity
with print skills, such as knowing that print reads left to right.41 In terms of concrete mathematics
knowledge, 94 percent of children entering kindergarten pass mathematics proficiency level one,
that is, reading numerals, recognizing shapes, and counting to 10.42
Finally, we review the factors that public school kindergarten teachers say are “very important” or
“essential” to kindergarten readiness—physical health and eagerness to approach new activities.43
Children’s health is reported as very good or excellent, with just 3 percent of children having “fair
or poor general health.” At the same time, 92 percent of children are “eager to learn.”44
Interestingly, only 10 percent of kindergarten teachers say knowing the letters of the alphabet is
very important or essential to being ready for kindergarten, and just 8 percent consider being able
to count as very important or essential.45
According to the proxy measures of preschool enrollment rates and early literacy activities,
concrete skills assessment at kindergarten entry, and measures ranked by kindergarten teachers as
important or essential to preparing children for kindergarten, most children entering kindergarten
appear to be equipped with the knowledge and traits required to begin the kindergarten year. The
apparently high levels of preparedness call into question the notion that there is a widespread need
for yet more government involvement in this arena.

Reason Foundation

Part 2

Full-day or Half-day? The Kindergarten
Decision

he state of Arizona provides a useful case study on the national trend toward full-day
kindergarten. In Arizona today, an estimated 56 percent of kindergartners attend half-day
programs, and 44 percent attend full-time.46 Currently, kindergarten is funded through diverse
sources including the state general fund, local taxes, and parent fees. Governor Napolitano has
proposed a centralized, statewide, full-day kindergarten program with a projected price of $200
million annually, not including current spending on kindergarten or an additional $100 million
required to build new classrooms.47
Will full-day kindergarten improve student achievement?

Local advocates point to the Alhambra and Chino school districts and the Reading First program as
evidence that full-day kindergarten is worthwhile. Testifying about her views of full-day
kindergarten, the governor said, “The Alhambra school district has long been a model for full day
kindergarten success.” The governor continued by citing a performance analysis conducted by the
Chino Valley unified school district, stating, “We know what works, we’re just not doing it.”48 Jim
Rice, Alhambra superintendent, believes the preschool programs are working, citing superior test
scores for students attending the district’s preschool programs. “This is the type of information to
get out to our Legislature,” Rice said. “This is working.”49
We examine summaries of those three programs and find their research designs of poor quality,
rendering them of little help in addressing the question of whether full-day kindergarten is
beneficial to students.
Campbell and Stanley’s classic 1963 work, Experimental and Quasi-Experimental Designs for
Research, has served as a basic text for social science researchers for generations, laying out a
variety of research designs and what the authors describe as “threats to validity.”50 Focusing on
education research, they explain eight internal threats to validity (in which a researcher mistakenly
attributes changes in an experimental group to the treatment), and four threats to external validity
(whereby the researcher cannot legitimately generalize the results of the experiment to broader
populations).

PRE-K EDUCATION
Only the highest form of experimental design, involving random assignment and a control group,
can hope to remove all 12 threats to validity. Through this technique, students are randomly
assigned to experimental and control groups, and if conducted properly, this technique creates two
groups that are nearly identical (within a measurable amount of random error)—making the
introduction of the treatment the crucial and measurable difference between the two groups.
Conversely, on the opposite end of sophistication, is a design known as “Static Group
Comparison.” The Alhambra and Chino summaries follow this design. Unfortunately, this design
controls for just four threats to validity, making it one of the least valid and informative designs.
The Reading First design is stronger, lacking only randomization, as explained below.
Alhambra. The Alhambra summary compares the scores of third-grade and fifth-grade students
who attended both preschool and full-day kindergarten to the scores of all third-grade and fifthgrade students in the district, and reports that scores are higher for children who attended preschool
and full-day kindergarten.51 The district concludes, “Students who attend a preschool program and
full-day kindergarten are better prepared and have a much greater chance of succeeding in
school.”52 However, it is unclear whether preschool and kindergarten attendance are responsible for
the difference in test scores.

Higher scores could be the result of family background or other student characteristics and

not due to preschool and all-day kindergarten.
Alhambra’s critical flaw is that the researchers did not test children before they entered the
programs, which means the differences between the two groups may have been present before the
children entered school. If for any reason—whether systematic or by random chance—those few
students began the program with higher scores than average, the study is without scientific value.
For instance, if the parents who placed their children in both preschool and all-day kindergarten
did so because they value education more highly than the average family, this could lead to
mistakenly attributing the higher scores of the treatment group to the program, when in fact the
higher scores could be either partially or wholly the result of family background or other student
characteristics. Given the body of research showing the primacy of family background and
influence as the strongest educational determinant, this oversight is critical.53 Without a pre-test or
random assignment, we simply cannot know whether the test score differences are a result of the
programs, family differences, self-selection bias, or other circumstances entirely.
Even if the findings are reflective of the Alhambra district, which cannot be discerned from the
data the district provided, the research design the district chose does not address external validity—
meaning that we can have no confidence that their results, even if accurate, can be generalized to
Arizona. To obtain this information, the researchers would have to measure various characteristics
of the student population, which was not done. The combined lack of pre-test information, random
assignment, and small sample sizes render the report of little value. Moreover, the Alhambra
summary is silent on the question of whether full-day kindergarten is more valuable than half-day

Reason Foundation
kindergarten since its treatment was preschool with full-day kindergarten: it simply did not control
or test for this information.
Chino. Chino shows the test scores of kindergarten students in one elementary school (Del Rio)
who had enrolled in full-day kindergarten compared to the scores of students who had enrolled in
half-day programs. It finds higher test scores for children in the full-day programs. It is unclear
from the summary how many children were tested each year, but the report states, “In the 2003
school year, out of 102 students in Del Rio’s kindergarten, parents of only 12 students chose a parttime program.”54 We assume the number of participating students was similar in the years tested.
The Chino summary suffers from the same flaws in the Alhambra report—no pre-test was
conducted to assess the children’s starting points, assignment to the programs was not random, and
the sample size was extremely susceptible to threats to validity. Lacking a pre-test, we simply
cannot know whether the test score differences existed prior to school entry. Lacking random
assignment, we cannot determine whether the test score differences are due to other factors, such as
the educational values or background of the families choosing one program over the other.
Additionally, with an assumed sample size of one dozen students, we can have no confidence that
the results are anything but random. In addition, children were not monitored past kindergarten.
The Chino data are therefore uninformative.
Reading First. The Reading First analysis has a stronger design than the Alhambra and Chino
summaries, yet it, too, suffers from important shortcomings. The report examines the test scores of
children in full-day and half-day kindergarten programs at school entry and at the end of the
kindergarten year, and finds that 59.8 percent of the full-day kindergarteners met the “benchmark”
compared to 42.6 percent of the half-day kindergarten group, a reported advantage of 17 percent.55
It finds the full-day group made more progress in reading than did the students in half-day
classes.56
Like the Alhambra and Chino summaries, the Reading First analysis is also susceptible to selection
bias, which means researchers cannot determine with any certainty whether the test score
differences are the result of the kindergarten programs, or whether the results may be due to other
factors such as the educational values of the families choosing one program over the other. Equally
problematic is the absence of controls on the background of the students. There is no multivariate
analysis, which would measure and control for a number of factors about each group. For example,
one might measure the family income of every child and the highest level of educational attainment
of each child’s mother and then run an analysis with each factor included as a separate control
variable. If, for example, the all-day kindergarten group had significantly higher family incomes
than the half-day group, it could lead to the impression that the all-day program led to score gains
when in fact it was a difference in family background that led to the appearance of an experimental
effect. Randomization minimizes these types of differences, and measuring and controlling for
them in a multivariate analysis could nearly eliminate them. The Reading First analysis does
neither. Moreover, the researchers did not measure whether the differences observed are
statistically significant (i.e., not likely to be the result of chance). Therefore it is a heroic
assumption to argue that all differences observed are due to the kindergarten programs.

PRE-K EDUCATION
Nonetheless, the size of the difference at the end of the kindergarten year is such that the data
might withstand the introduction of the proper controls and could be found to be significant. This
would be consistent with research showing that full-day kindergarten gives children a modest
academic advantage over children in half-day programs.57
We note, however, that the differences researchers observed in the Reading First analysis already
begin to fade by the beginning of first grade. At the end of kindergarten, 17 percent more of the
full-day students had attained the benchmark than those in the half-day program (59.8 percent
compared to 42.6 percent). As the Reading First analysis reported, just a few months later, at the
beginning of first grade, that advantage dropped almost in half, to 10 percent (58.7 percent
compared to 49 percent). Similarly, at the end of kindergarten, 15 percent more of the half-day
students were recommended for intensive support (34.7 percent compared to 19.9 percent), but by
the beginning of first grade, the difference had dropped to 10 percent (23.6 percent compared to
13.6 percent).

We can be sure that, if those advantages exist, they also fade quickly.
Therefore, the reasonable conclusion to draw from the Reading First analysis is that, while we
cannot be confident in the advantages of full-day kindergarten, we can be sure that, if those
advantages exist, they also fade quickly. This finding would be consistent with the highest quality
research conducted to date on kindergarten programs.
This is why the National Center for Education Statistics Early Childhood Longitudinal Study is so
important. As noted earlier, the researchers assessed 22,000 children at kindergarten entry and
most recently reported on those students through the third grade. The data set is the only one of its
kind, giving researchers information on dozens of variables that affect student achievement, and,
importantly, allowing them to control for the impact of kindergarten programs.
The ECLS-K research shows the same pattern documented by hundreds of early education studies:
children in full-day kindergarten are afforded a modest academic edge over children in half-day
kindergarten when measured at the end of the kindergarten year. However, that initial edge
completely disappears by third grade.
At the end of the kindergarten year, the researchers find there is “little meaningful difference” on
reading and math test scores between all-day and part-day kindergartners. They write, “In terms of
kindergarten program type (i.e., all day or part day), there is little meaningful difference in the
level of children’s end-of-year reading and mathematics knowledge.”58 What is the difference? “On
a reading scale that ranged from 0 to 72, the average kindergartner in a full-day program gained
10.6 points over the school year. For children in half-day kindergarten programs, the average gain
was 9.4 points.”59 Final reading scores were 32.1 and 31.3, respectively. The findings in
mathematics are parallel.60 The difference is modest, and all the more modest considering full-day
students spend twice as much time in school as their half-day peers.

Reason Foundation
Importantly, the “little meaningful difference” observed at the end of the kindergarten year no
longer exists by third grade. By the end of third grade, the researchers no longer detect a difference
between students who attended part-day or full-day programs. They write, “This report did not
detect any substantive differences in children’s third-grade achievement relative to the type of
kindergarten program (full-day vs. half-day) they attended.”61 The finding holds across all subject
matters tested. “Third-grade reading, mathematics, and science achievement did not differ
substantively by children’s sex or kindergarten program type.62
The NCES reports document on a large scale the piecemeal findings on early education that have
been trickling in for years: in the short term, more early education may confer more gains than
lesser amounts of early education, but over time, those advantages are not sustained. Unless or
until the elementary and secondary school system is improved, it is unlikely that preschool or
kindergarten will lead to a measurable improvement in school achievement.

What Impact Do Preschool and Kindergarten Have on Student Achievement? A
Historical Overview
The NCES findings are less surprising in historical context. From 1965 to the present day, the
United States underwent a sea change in formal early education. Preschool, then uncommon, is
now the mode.
As Figure 1 shows, only 5 percent of three-year-olds attended preschool in 1965—today, 42
percent attend. Sixteen percent of four-year-olds attended preschool in 1965—today, that figure is
68 percent. For five-year-olds, kindergarten has become almost universal.63

Figure 1: Percentage of Three- and Four-Year Olds Enrolled
in Pre-primary Programs, 1965-2003
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%

4- year-olds

3- year-olds
68%
42%

16%
5%
1965

2003

Source: National Center for Education Statistics, Digest of Education Statistics, 2004, Table 43, available at
http://nces.ed.gov/programs/digest/d04/tables/dt04_043.asp, nces.ed.gov/programs/digest/d03/tables/dt043.asp.

PRE-K EDUCATION
Despite the widespread use of formal early education programs, student achievement has shown
little to no improvement. For instance, Figure 2 shows fourth-grade reading, science, and math
scores on the National Assessment of Educational Progress (NAEP) have been little better than
stagnant since 1971, 1977, and 1978, respectively.
As noted author and education researcher Andrew Coulson reports, “Student achievement has
stagnated or fallen in most subjects since 1970…That is the verdict of the five most reliable
sources of evidence: the National Assessment of Educational Progress (NAEP), the International
Evaluation of Education Achievement (IEA), the Young Adult Literacy Survey (YALS), the
National Adult Literacy Survey (NALS), and the International Adult Literacy Survey (IALS).”64
Although the relationship between inputs and outcomes is more complicated than this linear
analysis suggests, if the proponents’ arguments are correct, we should expect to see at least some
relationship between the increased enrollment in early education programs and student
achievement. This is particularly true when the states have, over the same period of time, more
than tripled spending on education, increased teacher salaries, and reduced class sizes.65

Figure 2: Fourth-Grade Reading, Science and Math Scores on the NAEP, 1971–2004
300
Reading

250

Math

241
225
220

200

Science

229
219

208

150
1971-1978

2004

Source: 1971, 1977, and 1978 are the initial test years shown respectively in reading, science, and math. See U.S.
Department of Education, National Center for Education Statistics, “Results Over Time: NAEP 1999 Long-Term Trend
Summary Data Tables,” August 2000, available at nces.ed.gov/nationsreportcard/tables/Ltt1999/.
NAEP 2004 Long-term Trend Summary Data Tables http://nces.ed.gov/nationsreportcard/ltt/results2004/2004_sdts.asp

Certainly many factors contribute to student learning, but the lack of any apparent relationship
between increased enrollment in early education programs and later student achievement suggests
more formal early education is unlikely to improve student achievement.

Reason Foundation

Part 3

How Do U.S. Children Perform? An
International Examination

dvocates often point to France’s écoles maternelles as the ideal model for early childhood
education. According to Sandra Feldman, president emeritus of the American Federation of
Teachers, the United States “can’t afford not to” adopt a pre-primary program sculpted after the
coveted French system.66 Nearly all three- and four-year-olds in France are enrolled in center-based
institutions.67
Does the European model produce superior results?
If early education programs were essential building blocks for later school success, we would
expect European students to have a stronger showing than U.S. students on international tests,
particularly in the early elementary years. However, test scores reveal that U.S. students routinely
outperform their international counterparts in reading, math, and science in fourth grade—the
earliest year for which comparative test scores are available.
Figure 3 shows that U.S. fourth graders
demonstrate significantly better reading
literacy skills than their French peers.68

With a score of 542, U.S. fourth graders
also perform significantly better than the
international average of 500, and
outperform their counterparts in 26 of the
35 countries participating in the literacy
exam, including Germany and Italy, which
have enrollment rates similar to France.69
The top performance of U.S. readers was documented in an earlier version of the 2001 exam. On
the 1991 version, U.S. fourth graders surpassed students in France, East Germany, West Germany,
and Italy with significant margins.70

PRE-K EDUCATION

Figure 3: U.S. Fourth-Grade Reading Literacy Scores Exceed French Scores, 2001
600

550

542
525

500

450

400
U.S.

France

Source: U.S. Department of Education, National Center for Education Statistics, “International Comparisons in Fourth Grade
Reading Literacy: Findings from the Progress in International Reading Literacy Study (PIRLS) of 2001,” NCES 2003–073,
April 2003, 5.

Test data from the Third International Mathematics and Science Study show U.S. fourth graders
also have above-average math scores, and their science performance is third only to South Korea
and Japan.71 U.S. fourth graders earned a score of 545 in mathematics, performing significantly
better than the international average of 529, and surpassing their counterparts in 14 out of 26
participating countries.72 In science, U.S. fourth graders scored 565—far above the international
average of 524.
While U.S. fourth graders are “A” students on the international curve, that advantage does not last.
By eighth grade, U.S. student performance is slipping, and test performance is mediocre. As David
Hoff reported for Education Week, “In 1995, the nation’s fourth graders aced international
mathematics and science tests. By the time they reached the 8th grade in 1999, though, they had
become little better than C students on a global curve…”73 A similar decline occurs in reading.
Figure 4 shows U.S. fourth graders score higher than 70 percent of their international peers while
U.S. eighth graders perform little better than the international average.
Student performance continues declining, and by twelfth grade U.S. seniors are “D” students on
the international scale.74 Out of 21 countries tested in math and science literacy, U.S. twelfth
graders performed better than students in only three countries—Lithuania, Cyprus, and South
Africa.75 As the U.S. Department of Education describes it, “U.S. students performed relatively
well at the fourth-grade level, about average at the eighth-grade level, and below average at the
twelfth-grade level.”76 Figures 5 and 6 illustrate the decline.

Reason Foundation

Figure 4: Decline in U.S. Reading Literacy Performance from Fourth to Eighth Grade
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Fourth Grade

Eighth Grade

Source: Mullis et al., “PIRLS 2001 International Report: IEA’s Study of Reading Literacy Achievement in Primary Schools,”
Boston College, 2003, Chapter 1, available at timss.bc.edu/pirls2001i/pdf/P1_IR_Ch01.pdf; and U.S. Department of
Education, National Center for Education Statistics, “Outcomes of Learning: Results from the 2000 Program for International
Student Assessment of 15-year-olds in Reading, Mathematics and Science Literacy,” December 2001, Chapter 2, available
at nces.ed.gov/pubs2002/2002115.pdf.

PRE-K EDUCATION
Figure 5 shows U.S. students score higher than 58 percent of their international peers in the fourth
grade, but score higher than just 14 percent by twelfth grade. Figure 6 shows a similar decline in
science performance with U.S. students surpassing 92 percent of their international peers in fourth
grade, but performing better than only 29 percent by twelfth grade.

Figure 5: Decline in U.S. Math Performance by Grade Level

100%
80%
60%

58%
34%

40%

14%

20%
0%
Fourth Grade

Eighth Grade

Twelfth Grade

Sources: Fourth grade: “TIMSS highlights from the Primary Grades,” Boston College, June 1997: 2, Table 2, based on 1994–
1995 TIMMS data, available at isc.bc.edu/timss1995i/TIMSSPDF/P1HiLite.pdf. Eighth grade: “TIMSS Highlights from the
Middle School Years,” Boston College, November 1996: 2, Table 1, based on 1994–1995 TIMMS data, available at
isc.bc.edu/timss1995i/HiLightB.html. Twelfth grade: “TIMSS Highlights from the Final School Years,” Boston College,
February 1998: 2, Table 2, based on 1995–1996 TIMMS, available at isc.bc.edu/timss1995i/TIMSSPDF/C_Hilite.pdf.

Reason Foundation
What test scores reveal, then, is that U.S. students are strong competitors in the early elementary
years, excelling in reading and science and performing above average in math. Over time, U.S.
student performance declines and international students take the lead. Whatever the cause of that
decline, however, it appears to have little or nothing to do with a lack of preparation in the early
years. To the degree that international test data are informative, America’s decentralized and
flexible early education system is outperforming the European model and excels in equipping
students for superior achievement in the elementary years.

Figure 6: Decline in U.S. Science Performance by Grade Level
100%

92%

90%
80%
70%

61%

60%
50%
40%

29%

30%
20%
10%
0%
Fourth Grade

Eighth Grade

Twelfth Grade

PRE-K EDUCATION

Part 4

Examining the Economic Case for
Universal Preschool
A. Perry Preschool: Can $1 Today Yield $7 Tomorrow?
The Perry Preschool Project was a longitudinal experiment designed to study the effects of early
intervention on disadvantaged children. It was the early intervention program most frequently cited
in research reviews between 1983 and 1997, and is heavily cited in the literature and legislation in
support of universal preschool.77
Investigators at the High/Scope Educational Research Foundation in Ypsilante, Michigan
conducted the experiment from 1962 to 1965. The investigators reported their most recent findings
in “Lifetime Effects: The High/Scope Perry Preschool Study through Age 40.”78 The project was
an intervention program for three- and four-year-olds deemed at risk for “retarded intellectual
functioning and eventual school failure.”79 It involved either one or two years of half-day preschool
for seven months each year and periodic home visits. One hundred twenty-three children
participated, 58 children in the experimental group and 65 in the control group. All of the children
were of low socioeconomic status and had IQs in the range of 70 to 85.80 The study is frequently
cited because it is the longest running study of any preschool intervention program.
Analyses show that students who participated in the preschool program fared better over the long
term on a variety of educational and social measures than did children in the control group.
Lawrence J. Schweinhart, now president of the High/Scope Foundation, wrote, “Program
participation had positive effects on adult crime, earnings, wealth, welfare dependence, and
commitment to marriage.”81 On the basis of those findings, Schweinhart concluded, “The program
provided taxpayers a return on investment of $7.16 on the dollar.”82 Advocates rely heavily on that
figure to make their case that preschool is an investment that more than pays for itself in the long
term.
The High/Scope researchers’ interpretation of the long-term findings is that the preschool program
prepared children for kindergarten, which resulted in a more positive reaction by kindergarten
teachers that, in turn, caused the children to have a stronger commitment to school. That is
sometimes called the snowball hypothesis. Three researchers from Yale University explain,

Reason Foundation
The snowball hypothesis presumes that children who attend quality intervention programs are
better prepared socially and academically when they begin school. This enables them to
interact positively with their teachers, who in turn relate positively to them, and this tone of
adult-child relationships continues in progressive years of school.83
Others posit that the home visitation component was largely responsible for the results. They
hypothesize that people became more effective parents as a result of their involvement in the
program. Experiences such as building relationships with teachers may help parents establish a
more supportive home environment and effective “home-school linkages.”84 At any rate, there is no
consensus on which components of the program were responsible for the children’s gains. The
critical question remains: how could a one- or two-year half-day preschool program produce such
outstanding results?
The High/Scope researchers have been subject to heavy criticism for using nonstandard
significance levels. If standard significance levels are used, many of the most “significant”
differences between the experimental and control groups disappear.85 Psychology professor Charles
Locurto of the College of the Holy Cross in Massachusetts has argued that the Perry results are less
remarkable when all findings—not just those that favor Perry—are considered. Locurto writes,
We might marry the large number of nonsignificant and unfavorable findings into a different
picture of the Perry Project’s outcomes. We might argue that preschool training resulted in no
differences in school motivation or school potential at the time of school entry, no lasting
changes in IQ or achievement test performance….There were no differences in their average
grades as compared to former control-group children, in their personal satisfaction with their
school performance or in their self-esteem. Their parents were no more likely to talk with
teachers about school work or to attend school activities and functions than control-group
parents. Preschool children were more likely to have been placed in remedial education. By
age 19, they were unemployed at a rate equal to that of their control-group counterparts.86
More importantly, questions have been raised concerning the Perry sample and methodology.
According to Head Start co-founder and Sterling Professor Emeritus of Psychology at Yale
University, Ed Zigler,
[The Perry sample] was not only nonrepresentative of children in general; there is some doubt
that it was representative of even the bulk of economically disadvantaged children . . . The
Perry Project poses a number of methodological difficulties . . . Children had to have a parent
at home during the day, resulting in a significant difference between control and intervention
groups on the variable of maternal employment . . . [and] assignment to experimental and
control groups was not wholly random.87
Even if one believes the Perry findings are valid for disadvantaged children, they form a slippery
basis for universal preschool, and caution is in order. First, in more than 40 years, no other
program or study has produced results as dramatic as those found for Perry.88 That suggests that
there may have been unique conditions at the Perry Preschool that simply cannot be duplicated. As

PRE-K EDUCATION
a general principle, science requires an experiment to be replicable before it is considered valid.
Certainly caution is in order when it comes to applying findings to millions of children.
Second, benefits were obtained only for severely disadvantaged children at risk of “retarded
intellectual functioning.” It is simply inappropriate to generalize the
effects of Perry to mainstream children. This is particularly important given the research that shows
some early education programs do not always benefit—and may even be harmful to—mainstream
children.
Third, Perry children may have outperformed children in the control group, but they still fared
poorly compared with mainstream children. For example, nearly one-third of children participating
in the intensive program dropped out of high school; nearly one-third of the children were arrested;
and three of five participating children received welfare assistance as adults.89 That has led many
researchers to be more level-headed about the likely effects of early intervention: “Policymakers
should not assume that the widespread enrollment of low-income children and families in early
childhood programs will enable children living in poverty to perform later in school and life at the
levels reached by more advantaged [mainstream] children.”90
Finally, Perry differed significantly from regular preschool programs or what we could expect to
see in most universal preschool proposals. The fact that no other preschool program has ever
produced results akin to Perry may be testament to that.

B. RAND and The Chicago Child-Parent Centers
Much of the momentum for universal preschool in California and the nation comes from a RAND
Corporation study claiming that making universal preschool available in California would yield
$2.62 in benefits for every $1 spent.91 RAND bases these positive economic predictions for
California children on extrapolations from a study of a preschool intervention known as the
Chicago Child-Parent Center program. The RAND cost-benefit analyses suffer from many of the
same limitations as the Perry preschool study. The Chicago preschool program served the most
disadvantaged children in Chicago. These children were the subject of a longitudinal study of more
than 1500 disadvantaged children, about 1000 of whom went through the preschool program and
550 who did not. The study found children going through the program had less grade repetition,
less need for special education, higher graduation rates, less child abuse cases, and lower crime
incidents.92
However, comparing the Chicago program with the treatment that most universal preschool
programs offer is like comparing apples and oranges. The Chicago children had positive economic
outcomes because of intensive family and school interventions. For instance, the Chicago program
includes a parent program that includes a parent resource room with educational workshops,
reading groups and craft projects. Parents also volunteer in the classroom, attend school events
and field trips, and are assisted in completing high school.

Reason Foundation
The Chicago program also featured outreach activities including home visitations by staff. Also
health screening, speech therapy, and nursing and meal services were provided. In addition, many
of the children in the study received tutoring in reading and math until the third grade. As Arthur J.
Reynolds, the lead researcher in the Chicago study, stated in support of his findings, “We are
confident that participation in the Child-Parent Center (CPC) Program from ages 3 to 9 years was
the source of the group differences at age 20 years.”93 In contrast, most universal preschool plans,
including California’s “Preschool for All” initiative are small-scale interventions that would
involve children for one school year rather than six years.
The bottom line for the Chicago study is that it is difficult to say with certainty whether the
positive economic outcomes for the disadvantaged children were brought about by the preschool
intervention or the extra tutoring or the parental involvement. The RAND researchers were willing
to assign similar economic benefits for California children, even though the preschool treatment
would be entirely different than the treatment in the Chicago study.
Psychologist Dr. Matthew Thompson of Children’s Hospital in New Orleans, writing in the
American Medical Association Journal, critiqued the Chicago program, saying: “It is possible that
parental involvement explains more of the variance in outcome among inner-city children than do
structured programs. . . . If policy makers mistakenly accept the conclusion that preschool
intervention results in less criminal activity later, they may mistakenly invest in these programs
when the money might be better invested in parenting skill programs and other interventions to
increase parental involvement.”94
While the Chicago study offers at least some justification for assigning economic benefits to the
most disadvantaged children, RAND offers no justification for assigning economic benefits to
more–advantaged children that make up the majority of the children that would receive universal
preschool services in California and other states. According to former Legislative Analyst Bill
Hamm only 8 percent of funding from the proposed California program would go to enroll “highrisk” kids in preschool, who otherwise would not attend preschool.95
RAND’s study states that “there is little in the way of quantitative evidence to suggest how much
benefits would be attenuated [i.e., lessened] for more-advantaged children.” Yet, despite the lack
of this evidence, RAND arbitrarily makes the assumption that middle-income children would
receive 50 percent of the benefits of the Chicago program, while upper-income children would
receive 25 percent of the benefits. Yet there is no empirical justification to assign any economic
value to the benefits of universal preschool for more-advantaged children.

C. The Carolina Abecedarian Project
Although it is neither a preschool nor a kindergarten program, advocates often mention the
Abecedarian project because of its unique long-term findings. The Abecedarian Project was
launched in 1972 by researchers at the Frank Porter Graham Child Development Center in Chapel

PRE-K EDUCATION
Hill, North Carolina, and involved 111 children deemed at-risk on the basis of their parents’
income, education, and other factors. The mean age at entry into the program was 4.4 months. The
infants were placed in an eight-hour-a-day, five-day-per-week, year-round educational day care
center. They received free medical care, dietary supplements, and social service support for their
families. From ages five through eight, half of the children from both the experimental and the
control groups were given extra help in school and at home by specially trained teachers.96
At every age from one-and-a-half to four-and-a-half years, children treated in preschool
significantly outscored the control group on measures of intellectual development. At age eight,
test data showed significant positive effects of preschool treatment on intellectual test scores. A
follow-up test at age 12 showed that the effects of preschool treatment on children’s performance
on intellectual tests and on reading and mathematics tests had been maintained into early
adolescence. As the Abecedarian Project researchers note, “This represented a longer maintenance
of preschool intervention gains than has typically been reported from previous projects concerned
with similar children and families.”97 Most recently, researchers examined the children’s
intellectual and academic performance at age 21 and found that students who had received the
treatment “attained higher scores on both cognitive and academic tests, with moderate to large
treatment effect sizes.”98
As with the Perry project, there is no consensus on which components of the program were
responsible for the children’s gains, although it has been suggested that the early cognitive gains
were associated with greater mastery of academics, which led, in turn, to better performance
thereafter.99 The findings also provide support for the intensity or duration hypothesis, which
predicts that longer, more intense programs result in the most advantages for children.100
The Abecedarian Project has received some
criticism, most notably from Herman H. Spitz,
former director of the Research Department at the E.
R. Johnstone Training and Research Center in
Bordentown, New Jersey.101 Spitz expressed concern
that the project personnel presented certain results in
ways that bias the findings in favor of Abecedarian.
For example, by combining the IQ findings of the
four cohorts studied, the researchers concluded that
the intervention raised IQ. However, they neglected to report that scores improved only for two of
the four groups. In fact, for the third and fourth cohorts, the experimental group actually lost 3.68
IQ points more than did the control group, providing no support for the efficacy of the intervention
on this measure.102
Spitz also points out that differences favoring the intervention group first emerged at six months of
age, when those children’s advantage was six points. He writes, “This is a rather surprising finding
when one considers that the mean age of entry into the daycare center was 4.4 months.”103 The
intervention groups’ IQ advantage at five years of age was essentially the same as it had been at six

Reason Foundation
months of age. Spitz asks, “What happened during the initial 1.6 months to produce essentially the
same advantage for the intervention group that later was found at 5 and 12 years of age?”104 He
continues, “We need to understand why an additional 4.5 years of intensive intervention had so
little effect that, at six years of age (and older), the difference between the intervention and control
groups was not appreciably different than it had been at six months of age.”105 Spitz also argues
that because of the ways the tests were conducted, some of the reported test results may be biased
in favor of the Abecedarian Project.106
Whether or not one takes the Abecedarian findings as wholly valid, there are several facts that
should prevent legislators from basing policy recommendations for universal early education on
the study. First, the Abecedarian project did not include mainstream students, and benefits were
obtained only for a small group of “economically disadvantaged African-American children.” The
findings do not inform questions regarding mainstream children.
Second, Abecedarian was not a one-, two- or even three-year preschool or kindergarten
intervention. It was an intensive intervention that created a home-away-from-home for infants and
continued at an intensive level for more than five years. It was not akin to preschool or
kindergarten programs. It was a full-time intervention from birth through age five that few parents
would find comfortable.
Finally, the Abecedarian Project was the first of its kind and has not been repeated. As the authors
report, “The persistence into adulthood of the Abecedarian treatment effects on cognitive
development is in contrast to the erosion of treatment/control test score differences in the Early
Training Project and the High/Scope Perry Preschool Project, the only other randomized trials of
early childhood intervention to have reported post-high school findings.”107 Because the
Abecedarian Project was the first of its kind to demonstrate sustained results, it is important that it
be replicated—and the factors leading to such anomalous findings are understood—before drawing
further conclusions.
Whatever their merits, neither Perry Preschool, the Chicago Child Parent Centers, nor Abecedarian
speaks to mainstream children nor to the type of preschool or kindergarten programs proposed by
today’s policymakers. Additionally, all were model projects that treated a small group of children
in specific conditions. Could those effects be expected of widespread public programs? On this
point, information on Head Start is informative. Head Start is the government’s longest running
preschool program for disadvantaged children and it has failed to produce long-term academic
advantages for participants.

D. Head Start
Research on Head Start is valuable because it is a large program operating under real-world
conditions and constraints, and research has been conducted over a 40-year period. Head Start has
more than 1,300 preschool projects serving about 457,000 disadvantaged children. The information

PRE-K EDUCATION
about the effects of Head Start can serve as a close approximation of what one might expect from a
universal preschool education for disadvantaged children.
Like many of today’s early education advocates, former president Lyndon B. Johnson sold his
program to the public by promising that early intervention could prevent delinquency, poverty, and
welfare use.108 The reality of Head Start has been much different. Head Start programs have had
mixed short-term results. Consistent with broad findings on early education, however, Head Start
students have not demonstrated lasting achievement gains.
In 1997, the General Accounting Office (GAO) conducted a thorough analysis of Head Start’s
impact.109 After speaking with early childhood researchers and practitioners and searching through
electronic databases to locate published and unpublished manuscripts, the GAO found nearly 600
citations and documents. Of those, only 22 studies fit their criteria for review, and even those “had
some methodological problems.”110 No study used a nationally representative sample so that
findings could be generalized to the national program. The GAO concluded, “The body of research
on current Head Start is insufficient to draw conclusions about the impact of the national
program.”111

Head Start students have not demonstrated lasting achievement gains.
The Department of Health and Human Services (HHS) has maintained that research proves Head
Start’s effectiveness. In a letter to the GAO, June Gibbs Brown, then inspector general of HHS,
wrote, “There is clear evidence of the positive impacts of Head Start services.”112 For supporting
evidence, HHS cited findings from a comprehensive synthesis of Head Start impact studies
conducted under HHS auspices in 1985.113 The study showed that Head Start could have an
immediate positive impact on cognitive measures, social behavior, and child health. HHS neglected
to mention the rest of the findings—namely that the short-term impact of Head Start diminished
once the children entered school. The synthesis concludes, “In the long run, cognitive and
socioemotional test scores of former Head Start students do not remain superior to those of
disadvantaged children who did not attend Head Start.”114
On the three cognitive measures tested (IQ scores,
school readiness, and achievement test scores), the
report found, “Once the children enter school there
is little difference between the scores of Head Start
and control children … Findings for the individual
cognitive measures—intelligence, readiness and
achievement—reflect the same trends as the global
measure…By the end of the second year there are
no educationally meaningful differences on any of
the measures.”115

Reason Foundation
Findings on children’s social behavior, achievement motivation, and self-esteem were similar: “On
social behavior, former Head Start enrollees…drop to the level of comparison children by the end
of the third year. On achievement motivation and self-esteem, Head Start children drop below nonHead Starters a year after Head Start, then score about the same as comparison children for the
next two years.”116
In 2003, researchers released a new study on Head Start with a nationally representative sample of
2,800 children in 43 different Head Start programs called “Head Start FACES 2000.” The report
seems to confirm earlier findings. The researchers report, “Despite the gains they make, Head Start
children enter Kindergarten still substantially below national averages on such assessments.”117
Longer-term assessments have not yet been conducted, but are currently underway by Westat.

PRE-K EDUCATION

Part 5

Evaluating Universal Preschool in
Georgia and Oklahoma
A. Experience in Georgia: $1.5 billion on Universal Preschool Bears No Fruit
In 1993, the Georgia state legislature established a no-fee pre-kindergarten program now serving
an estimated 63,000 four-year-old preschoolers. Using the Georgia Kindergarten Assessment
Program (GKAP), in 1999 researchers at Georgia State University tested children who had
participated in the preschool program and compared their scores to all students in the state during
the kindergarten year. Both groups scored well, but their scores were indistinguishable.
The researchers concluded, “Eighty-eight percent of the study sample scored a five on the
capability item, compared to 85 percent of all students across the state scoring similarly. Statistical
tests indicate that overall these differences are not significant. In other words, the study sample
does not differ from the entire kindergarten population in GKAP capability scores.”118
Reports also show that GKAP scores are essentially the same as they were before Georgia adopted
the universal preschool program. Linda Schrenko, then Georgia State School Superintendent,
expressed the state’s disappointment, saying, “The only message you can get from it is that our
kindergarten non-ready rate [7 percent of students] is the same, regardless of what we do.”119
In 2003, Georgia State University researchers released the latest findings from the fifth and final
year of the longitudinal study of the pre-kindergarten program. In the final report, they write,
“Previous research has shown that cognitive gains as measured by standardized test scores are
associated with preschool experiences but are not sustained in later years…It should not be
surprising to find that the test scores of children, all of whom participated in a pre-k program four
years before are not systematically different.”120 The researchers show the test scores of children
who remained on grade level and who were not exempted from state testing by virtue of their
individualized education plan and report their average percentile test scores in math, language arts,
science and social studies: all fall below the national average and are not systematically different
from Georgia’s average student performance.121

Reason Foundation
Other findings on grade retention and curriculum are also informative. The researchers report,
“About 15 percent of the children were retained at least once by their fourth year of primary
school.”122 Within the preschool control group, researchers were also able to assess the impact of
varying types of preschool curriculum and found, “Students’ economic backgrounds have more
influence on their educational success after pre-k than curriculum choice and teacher
credentials.”123 Lead researcher Gary Henry writes, “Program characteristics made only small
differences in retention and test scores. These differences are much less dramatic than some of the
differences based on parental education or socio-economic status…There is no magic bullet here.
No one thing is waiting in the wings to increase scores for all students…”124
After ten years, the Georgia preschool program has served over 300,000 children at a cost of $1.15
billion and children’s test scores are unchanged.125 Because programs like these, once in place and
supported by special interests, are virtually never removed, Georgia’s taxpayers are still paying the
bill and probably will continue to do so.

B. The NAEP and Universal Preschool
Evidence from performance on the National Assessment of Education Progress (NAEP), which is
considered the nation’s report card, also calls into question the value of investing in universal
preschool. Georgia has had universal preschool open to all children since 1995 and Oklahoma has
had a universal program in place since 1998. Yet, the overall performance of these states on the
NAEP in terms of reading achievement calls into question the lasting value of universal preschool
on academic outcomes. In a recent analysis of the top 10 best and worst state performers, based on

PRE-K EDUCATION
the percentage point change in fourth-grade reading tests between 1992 and 2005 on the NAEP,
both Georgia and Oklahoma were in the bottom 10 performers.126 In fact, Oklahoma was the worst
performer of all states in terms of gains in fourth-grade reading between 1992 and 2005, actually
losing 4 percentage points.
More specifically, in Oklahoma 33 percent of fourth graders were below basic in reading in 1992.
By 2005 40 percent of Oklahoma fourth graders were scoring below basic. In 1992, 38 percent of
Oklahoma fourth graders scored basic in reading. Again by 2005 only 35 percent of fourth graders
could read at a basic level. Finally, in 1992, 25 percent of Oklahoma fourth graders were
proficient in reading, and by 2005 only 21 percent of Oklahoma fourth graders were proficient in
reading. (see figure 7)
One would expect that a large statewide investment in universal preschool, including highly paid,
credentialed teachers and a high-quality curriculum would have a positive effect on fourth-grade
reading scores. These scores declined, despite the fact that all of the children that took the 2005
NAEP reading test in Georgia and Oklahoma were eligible for universal preschool.
On the other hand, none of the states in the top 10 best performers in terms of gains in fourth-grade
reading on the NAEP between 1992 and 2005 had implemented universal preschool.

Figure 7: Title: Oklahoma Reading Achievement on NAEP, 1992-2005

Below basic

Basic

Proficient

Advanced

100%
80%
60%
40%
20%
0%
1992

Oklahoma

2005

1992

Source: Quality Counts at 10: A Decade of Standards-Based Reform, Education Week, 2006.
http://www.edweek.org/media/ew/qc/2006/17shr.ok.h25.pdf

U.S.

2005

Reason Foundation

Part 6

Recommendations
A. Increase Transparency
We recommend that state and federal policymakers bring transparency to current early education
spending by identifying, documenting, and tracking the amount of federal and state spending on
child care, preschool, and kindergarten programs in their jurisdictions before expanding
government subsidies for early childhood education.
Many state and federal early care programs have a poor track record of financial accountability.
The quality of financial data about early childhood education programs is often poor and difficult
to evaluate. Arizona is a case in point. Arizona State University (ASU) researchers compiled “The
Condition of Early Childhood Education and Care in Arizona: 2004.” They concluded, “The data
on early childhood education and care (ECEC) in Arizona are poor,” and recommended that the
School Readiness Board be given “the authority and funding to identify and track annually the
amount of federal and state dollars invested in ECEC.”127
In addition to poor quality data, fraud and mismanagement have plagued many large-scale
government-funded child-care programs. In its 40-year history, Head Start has proven to be rife
with financial abuse, mismanagement, and outright theft. A 2005 investigation by the House
Committee on Education found that because of fraud and poor management, a significant share of
the nearly $7 billion being invested annually by American taxpayers in Head Start never reaches or
benefits disadvantaged children.128
In New Jersey, where universal preschool has been implemented by court degree in 30 low-income
Abbott districts, about a dozen state-contracted child-care centers—nearly a third of the number
audited by the state—have been referred to the state attorney general’s office for possible
prosecution.129 Instead of educating children, administrators bought cars and jewelry, paid
themselves six-figure salaries, and even went gambling in Atlantic City. Researchers also found a
wide gap in quality among preschool classrooms as they sat in on hundreds of classes across the
state to evaluate New Jersey’s $400-million-a-year effort to help needy children catch up to their
wealthier suburban peers.130 They found significant problems in many of New Jersey’s state-

PRE-K EDUCATION
funded classrooms for poor preschoolers, including harsh discipline, a dearth of books, and
weaknesses in science and math.
Similarly, California currently spends more than $3 billion a year on subsidized preschool for lowincome children. A recent report by the District Attorney in charge of welfare fraud in California
reports that rampant fraud is costing California taxpayers as much as $1.5 billion a year—half of
the welfare money it pays to needy families for child care.131 In Los Angeles, for example, officials
estimate Los Angeles County loses 40 to 50 percent of its $600 million-a-year child-care allocation
to fraud.132
We recommend that an independent body, such as Arizona’s auditor general or California’s state
auditor, carefully evaluate current spending on early education and that any early childhood
program maintain more transparency to continue receiving public dollars.

B. Streamline Existing Duplicative Early Childhood Programs
Most states have several administrative agencies responsible for early childhood education funding
and program administration. For example, California currently spends more than $3 billion a year
on subsidized preschool for low-income children through 20 different funding streams. Rather than
creating yet another preschool bureaucracy and tax-funded revenue stream, California can create a
single, integrated, seamless administrative system that will serve low-income families. The
different funding streams that support low-income families have multiple administrative
bureaucracies, paperwork requirements, and eligibility requirements. Millions of dollars that could
go directly to families to pay for more low-income preschool slots are wasted maintaining
duplicative preschool programs. States need to streamline existing early childhood programs and
work toward one-stop provision with a centralized eligibility list for disadvantaged preschoolers.

C. Assess Impact
As a matter of good public policy, we recommend that state legislatures require an impact
assessment of early education expenditures. This is particularly important in light of empirical
evidence demonstrating the inability of early education programs to improve academic
performance. We concur with Arizona State University findings that, “The variety of agencies and
groups involved and the lack of a systematic and coordinated statewide data plan make it difficult
to evaluate the validity, integrity, and consistency of the ECEC available data…Pre-school and
kindergarten enrollment data are not systematically collected or organized…”133 There is almost no
information available on the impact of the more than $410 million spent annually on these
programs on student learning.134 This assessment applies to most early education programs and
evaluation of full-day kindergarten and universal preschool programs are a crucial component to
any large-scale early childhood funding expansion.

Reason Foundation

D. Transform Current Spending into Grants
To the degree that the states continue to be involved in early education, we recommend funding be
modified into direct education grants to families. The arguments for a flexible funding system of
per-child grants have been made extensively elsewhere.135
For example, imagine if Arizona’s early childcare system were redesigned so that funding followed
the child through direct grants. The best available estimates show that Arizona currently spends
more than $410 million annually on early education programs.136 Of that, an estimated $134
million is spent on kindergarten, a conservative figure that does not include the multiple funding
streams being utilized by school districts to provide kindergarten or parent fees.137 Nonetheless, the
amount is sufficient to give a kindergarten grant worth $1,950 to every kindergartner in Arizona.138
If grants were targeted to kindergartners whose family incomes were equal to or less than 185
percent of the federal poverty line (an estimated 42 percent of kindergartners), grants could be an
estimated $4,650 per child.139
The best estimates also show Arizona spends $265 million annually on non-kindergarten early
education programs, an amount sufficient to give every four-year-old in Arizona a grant of
$3,460.140 If grants were restricted to children whose family incomes were equal to or less than 185
percent of the federal poverty level, current spending could transform into grants in the estimated
amount of $8,240 per child.141
These figures are not intended to be prescriptive or definitive, but rather to illustrate the amount of
money currently spent on these early education programs and the possibility of using those funds
more efficiently through per-child grants.
At current spending levels, the grant amounts could be sufficient to assist parents’ choice among a
range of private providers. A recent Goldwater Institute survey of Arizona’s private schools finds
the average tuition for private elementary schools is an estimated $3,700.142 This figure is aligned
with national figures from the U.S. Department of Education that report the average private
elementary school tuition is $3,267.143
Absent a grant program that uses the private sector, the further provision of state-run early
education is likely to inflate costs. The California Preschool for All initiative estimates the cost of a
three-hour, 180-day program to be $8,000 per child. The Arizona School Readiness Task Force
says quality preschool costs at least $7,000 to $10,000 per child.144 Similarly, the National Institute
for Early Education Research estimates the per-child cost at $8,700.145 As with the provision of K–
12 education, publicly run preschool and kindergarten will likely cost significantly more than
privately provided options.
Quebec’s experience with seven years of universal preschool provides a cautionary tale for the
high costs of state-run preschool. The program that was supposed to cost $235 million over five
years now gobbles $1.7 billion every year.146 Yet there are not enough day cares to go around,

PRE-K EDUCATION
forcing Quebec parents to put kids that have not even been conceived on a waiting list. (A
Canadian likened getting a space in a day care to winning a lottery.) Half of the day care spaces are
taken by the top 30 percent income bracket. Currently there are still about 35,000 names waiting to
take advantage of the unbeatable daily deal.
In fact, Canada appears to be moving away from state-regulated early care toward a system of
grants to families. The new minority Conservative government plans to cancel funding of stateregulated early childhood education programs worth $3.7 billion. The money was promised to the
provinces and territories by the former Liberal government to help finance regulated, affordable
child-care spaces with an educational component.
The government is scrapping the Liberal program in favor of a taxable allowance of $1,200 a year
for each child under age six. Human Resources Minister Diane Finley says the provision will be
contained in the first Conservative budget, expected in late April or early May, and the first
monthly checks of $100 per child could go out in July 2006.147
Making use of private providers will also help reduce construction costs, which are projected at
$100 million for Governor Napolitano’s kindergarten plan.148 Similarly, in California the universal
preschool initiative includes more than $2 billion for new public preschool facilities. The practical
approach of making use of the private sector was a key factor in Florida’s recent decision to
implement preschool through a grant system.149 On January 2, 2005, Florida Gov. Jeb Bush signed
a bill allowing Florida’s four-year-olds to attend a preschool program of their parents’ choice—
including private centers. The per-child cost is expected to be between $2,000 and $3,000.150
Alex Penelas, the democratic mayor of Miami-Dade County who championed the initiative, said he
was working all the time under the assumption that parents would be able to choose either a public
or private school for their children, saying, “That’s more a practicality of having 90,000 children
arrive on the doorstep.”151 Author of the Florida Senate bill creating the program, Senator Lisa
Carlton (R–Sarasota), concurred, saying, “Because we don’t have enough spaces in the public
schools, it’s necessary for Florida to partner with the private sector.”152
According to the first program evaluation of Florida’s universal preschool program, 92 percent of
parents surveyed reported that they were satisfied with the program and 88 percent reported that
their child was learning reading and math.153 The competition between preschools to attract new
state-funded preschoolers has also led to improvements in school quality. The number of Florida
early childhood programs accredited by the National Association for the Education of Young
Children increased from 781 to 879 in the past year.154
We recommend a flexible system akin to Florida’s, allowing parents to spend their grants in any
public or private preschool or kindergarten program of their choice. Policymakers should ensure
the continued independence of private providers. This will allow families to choose from a
diversity of curricula, hours, and standards that suit individual student learning needs

Reason Foundation
A tax-credit approach could also help policymakers achieve the policy goal of more quality
preschool for children with more efficiency for taxpayers and greater satisfaction for parents.155
By supporting new preschool slots for low-income and middle-class children, all taxpayers would
be able to keep more of their own income to pay for their own preschool choices. A $1,000 taxcredit to middle-income families would help them to choose from a wider preschool market, and a
corporate tax-credit scholarship program could be created to give scholarships that would enable
low-income children to attend existing preschools. Pennsylvania’s example of the corporate
program shows that companies have been responsive to tax incentives. The state expanded the
existing K-12 corporate tax-credit program in 2003, giving corporations a 100 percent credit for
the first $10,000 and up to a 90 percent credit for remaining contributions up to $100,000. To date,
$5 million a year is used to target Pennsylvania’s low-income children with preschool scholarships.
Families of children receiving the scholarships must earn less than $50,000 plus a $10,000
allowance for each dependent. In the first year of the program, 39 preschool scholarship
organizations were created.156
New state-run programs may also threaten the private and parochial provision of services, and with
them, the diversity that is critical to meeting student needs. California’s Preschool for All Act
explicitly requires non-religious preschool provision. England’s experience is instructive in this
regard. The Department for Education and Employment worked vigorously to provide free
preschool places for all four-year-olds and most three-year-olds by 2002. The BBC news reported,
“The developments have proved disastrous for the private and voluntary sector.”157 More than
2,000 groups have closed since 1997, and 1,500 avoided closure only because of emergency
funding from the government.158
A new analysis on preschool competition between the private and public sector in Georgia
demonstrates the value of competition between the public and private preschool sector. The study,
Competition in the Sandbox, found that children who attended community-based preschools
displayed stronger language development and lower rates of grade retention compared to children
who attended preschools in the public schools.159 In addition, the study found that greater
competition improved educational outcomes for children in both public and private preschools.
Policymakers have the opportunity to transform current expenditures into a flexible system that can
provide for a more cost-effective use of funds, greater choice for parents, and a wider range of
opportunity for students.

PRE-K EDUCATION

Part 7

Conclusion

ov. Janet Napolitano has argued that “Today in America, we are trying to prepare students for
a high-tech world of constant change, but we are doing so by putting them through a school
system designed in the early 20th century that has not seen substantial change in 30 years.”160 We
agree, and elsewhere have argued for fundamentally changing the school system through the
powerful mechanism of school choice.161 Yet, Arizona’s governor has proposed an expansion of
the status quo. And several states including California and Illinois are moving to expand state-run
preschool.
To the degree that the state remains involved in early education, we recommend adopting a flexible
system of per-child grants. For example, current state spending on kindergarten is sufficient to give
a kindergarten grant worth $1,950 to every kindergartner in Arizona. We note this is a conservative
estimate that does not include the multiple sources of revenue currently generated by school
districts or parent fees, which could also be used to augment the amount. If grants were targeted to
children in families of modest means, kindergarten grants could be an estimated $4,650 per
student.
Empirical evidence suggests more early education will do little to improve children’s long-term
education outcomes. We summarize some key findings here:

The National Center for Education Statistics’ longitudinal study of 22,000 children finds no
lasting reading, math, or science achievement differences between children who attend halfday and full-day kindergarten. “This report did not detect any substantive differences in
children’s third-grade achievement relative to the type of kindergarten program (full-day vs.
half-day) they attended.”162

Georgia’s universal preschool program has not improved children’s academic performance.
“The study sample does not differ from the entire kindergarten population in GKAP capability
scores.”163

Georgia and Oklahoma are in the bottom 10 worst performers for reading achievement on the
NAEP fourth-grade reading assessment in 2005, despite years of investment in universal
preschool.

Head Start, the nation’s largest preschool program for disadvantaged children, has not
measurably improved educational outcomes. “Once the children enter school there is little

Reason Foundation
difference between the scores of Head Start and control children…Findings for the individual
cognitive measures—intelligence, readiness and achievement—reflect the same trends as the
global measure…By the end of the second year there are no educationally meaningful
differences on any of the measures.”164

Historic trends are not encouraging. The preschool enrollment rate of four-year-olds has
climbed from 16 percent to 66 percent since 1965. Despite the sea change from home
education to formal early education, we find student achievement has stagnated in most
subjects since 1970.

The French model of early education is not encouraging. French students have significantly
lower literacy rates than U.S. students as measured in fourth grade, the earliest year for which
comparative data are available.

America’s flexible approach to early education gives children a strong foundation, according to
widely used proxy measures of preparedness, concrete skills assessments, and reports by
kindergarten teachers. We find further evidence of the strength of our early education system
in international comparisons, which show U.S. fourth graders are “A” students on the
international curve, excelling in reading and science and performing above average in math.

By twelfth grade, U.S. students are “D” students on the international scale, a decline occurring
after fourth grade that implicitly must be addressed through reform of the current system.

For these reasons, among others, we strongly recommend against “ensconcing early care and
education as a lockstep component of public schooling,” and recommend alternative measures for
improving early childhood education and ultimately K-12 schooling—including transparency,
impact assessment, and individual student funding.

PRE-K EDUCATION

About the Authors

arcy Olsen is President of the Goldwater Institute. Under her leadership, Goldwater has
become Arizona’s most frequently cited research organization, working to expand school
choice, restore economic liberty, and restrain the growth of local and state government.

Before heading Goldwater, Olsen served as director of education policy at the Washington D.C.based Cato Institute. An authority on early childhood education, primary and secondary education
reforms, and the marketplace for education services, she has been a guest on local and national
media programs such as National Public Radio, the O’Reilly Factor, the Lehrer News Hour, and
Arizona’s Horizon. Her research findings have been widely published in such outlets as the Wall
Street Journal, the Washington Post, and USA Today, and she has testified before Congress. Before
joining Cato, she served as a transitional house manager for the D.C. Coalition for the Homeless
while earning a bachelor’s degree from the School of Foreign Service at Georgetown University.
Her master’s degree is in international education from New York University.
Lisa Snell directs the Education Program at Reason Foundation, where she oversees research on
education issues.
Ms. Snell, who has testified before the California State Legislature and numerous government
agencies, has recently authored policy studies on universal preschool, school violence, charter
schools, and child advocacy centers.
Ms. Snell is a frequent contributor to Reason magazine. She is regular contributor to School
Reform News and Privatization Watch. Her commentary has also appeared in Wall Street Journal,
USA Today, San Francisco Chronicle, the Orange County Register, Los Angeles Daily News, and
numerous other publications.
Ms. Snell is also the President of the Board of the California Virtual Academy, which is a K12
virtual charter school.
Before joining Reason Foundation, Ms. Snell taught public speaking and argumentation courses at
California State University, Fullerton.

Reason Foundation

Related Reason Foundation Studies
Violence Prevention: Strategies to Keep Schools Safe, by Alexander Volokh and Lisa Snell, Policy
Study #234, October 1997, www.reason.org/ps234.pdf
Special Education Accountability: Structural Reform to Help Charter Schools Make the Grade, by
Lisa Snell, Policy Study #319, July 2004, www.reason.org/ps319.pdf
School Vouchers as a Legal Sanction, by Lisa Snell, Policy Study #284, July 2001,
www.reason.org/ps284.pdf
Response to Bush Education Plan: School Vouchers are Needed to Enforce Bush’s Accountability
Plan, by Lisa Snell, E-brief #110, January 2001, www.reason.org/ebrief110.html
Reforming Remedial Education: Private Alternatives to Traditional Title I, by Lisa Snell with
Lindsay Anderson, Policy Study #266, January 2000, www.reason.org/education/ps266.html

PRE-K EDUCATION

Endnotes
1

Cynthia G. Brown, Elena Rocha, and Amanda Sharkey, Getting Smarter, Becoming Fairer: A
Progressive Education Agenda for a Stronger Nation, The Center for American Progress,
August 2005.

“Virginia Governor Tim Kaine’s Response,” The Washington Post, Transcript, January 31,
2006.

“Most States Increased Pre-K Funds this Year,” Report on Preschool Programs, vol.37, no. 22,
November 23, 2005.

Julie Poppe and Steffanie Clothier, “The Preschool Promise: Going to Preschool Benefits
Children Their Entire Lives. Can States Afford It to All Kids?” State Legislatures, June 2005,
pp. 26-27.

Full Day Kindergarten Legislation, AB 2046,
http://www.universalpreschool.com/alerts/20060410.asp

Full-Day Kindergarten: A Study of State Policies in the United States, Education Commission
of the States, June 2005. http://www.ecs.org/clearinghouse/62/41/6241.pdf

See “Preschool for All Act,” as filed with California Office of the Attorney General, June 20,
2005. http://www.earlyeducation.org/pfa1.pdf

Laura Rosenhall, “Reiner Revisits Preschool Initiative,” Sacramento Bee, April 19, 2005.

Janet Napolitano, “Governor Napolitano’s School Readiness Action Plan,” Office of the
Governor, State of Arizona, January 22, 2004: pp.3, 7.

Janet Napolitano, “Remarks to Supporters of the National Task Force on Public Education,”
Mayflower Hotel, Washington, D.C., April 22, 2004: p.4.

Napolitano, “Governor Napolitano’s School Readiness Action Plan,” p.2.

The exception is the United Kingdom, where enrollment of three-year-olds is 52 percent. U.S.
Department of Education, National Center for Education Statistics, “Comparative Indicators of
Education in the United States and Other G-8 Countries: 2002,” NCES 2003–026, May 2003:
26. Also available at nces.ed.gov/pubs2003/2003026.pdf (accessed January 18, 2005).

W. Steven Barnett et al., “The State of Preschool: 2004 State Preschool Yearbook,” National
Institute for Early Education Research, 2004, available at nieer.org/yearbook/pdf/yearbook.pdf
(accessed January 18, 2005).

National Childcare Information Center, Federal and State Spending for Early Care and
Education, http://www.nccic.org/poptopics/ecarefunding.html, accessed April 15, 2006.

“Short term” is defined as one to three years after program participation. For further reading,
see Arthur J. Reynolds et al., “The State of Early Childhood Intervention: Effectiveness,

Reason Foundation

Myths and Realities, New Directions,” Focus 1, 19 (Summer/Fall 1997): p.6. See also Donna
Bryant and Kelly Maxwell, “The Effectiveness of Early Intervention for Disadvantaged
Children,” in The Effectiveness of Early Intervention, Michael J. Guralnick, ed. (Baltimore,
Md: Paul H. Brookes Publishing Co., 1997): pp.23–46.
16

The questions and answers in this survey reflect views on child-care systems. However, child
advocates regularly use the term “early childhood care and education,” which reflects the belief
that child care and education are considered joint—not separate—functions. The survey is
informative about parents’ and advocates’ basic beliefs regarding the appropriate role of
government in the child care and early education arena. Public Agenda, “Necessary
Compromises: How Parents, Employers and Children’s Advocates View Child Care Today,”
Public Agenda, 2000:pp.11–13.

See Darcy Olsen, “Mandatory preschool another power grab,” Milwaukee Journal-Sentinel,
August 2, 2001, available at www.cato.org/dailys/08-14-01.html (accessed January 18, 2005).

Zell Miller, cited in Jodie Morse, “Preschool for Everyone,” Time, November 9, 1998.

Vermont State Legislator Bill Suchmann passed away in 2001. See “VT GOP Mourns Passing
of Former Representative Bill Suchmann,” available at www.vermontgop.org/suchmann.shtml
(accessed January 18, 2005). See also Anne Geggis, “Mandatory preschool?” Burlington Free
Press, February 16, 1998.

This paper does not examine the day care debate. For analysis on the availability, affordability,
and quality of child-care arrangements, see Darcy Olsen, “The Advancing Nanny State: Why
the Government Should Stay Out of Child Care,” Cato Institute Policy Analysis no. 285,
October 23, 1997; Robert Rector, “Facts about American Families and Day Care,” Heritage
Foundation FYI no. 170, January 21, 1998; Linda Giannarelli, Sarah Adelman, and Stefanie
Schmidt, “Getting Help With Child Care Expenses,” Urban Institute, Occasional Paper no. 62,
February 2003; and Arizona Department of Economic Security, “Child Care Market Rate
Survey 2000,” August 2000.

Tom Horne, “Our Views on All-Day Kindergarten,” Arizona Republic, November 21, 2004, v.
2.

Janet Napolitano, Message of the Week, February 18, 2004.

State School Readiness Board, “Summary of Priority Recommendations,” November 18, 2003:
p.5.

Jack O’Connell, “Why Quality Preschool Matters,” San Francisco Chronicle, February 13,
2006.

Edward Zigler, Cara Taussig, and Kathryn Black, “Early Childhood Intervention: A Promising
Preventative for Juvenile Delinquency,” American Psychologist 47, no. 8 (August 1992):
p.1000.

This discussion does not consider the health or nutrition components of early intervention
programs. “Long term” is defined as four or more years after program participation. Most
studies of this kind are impaired by high attrition rates and selection bias and are limited in
statistical power and generalizability. For reviews, see Walter Gilliam and Ed Zigler, “A
Critical Meta-Analysis of All Evaluations of State-Funded Preschool from 1977 to 1998:
Implications for Policy, Service Delivery and Program Evaluation,” Early Childhood Research
Quarterly 15, no. 4 (Winter 2000): pp.441–473; Reynolds et al., “The State of Early
Childhood Intervention: Effectiveness, Myths and Realities, New Directions,” p.6; Bryant and

PRE-K EDUCATION

Maxwell, “The Effectiveness of Early Intervention for Disadvantaged Children,” pp.23–46;
and Nancy L. Karweit, “Effective Preschool Programs for Students at Risk,” in Effective
Programs for Students at Risk, R. E. Slavin, ed. (Needham, MA: Allyn and Bacon, 1989):
pp.75–102.
27

Russell W. Rumberger and Loan Tran, Preschool Participation and the Cognitive and Social
Development of Language Minority Students, UC LMRI Technical Report, University of
California, Santa Barbara, January 2006. http://lmri.ucsb.edu/publications/06_rumbergertran.pdf.

David Weikart cited in Kenneth T. Walsh, “The three R’s and the big P,” U.S. News & World
Report, August 30, 1999, p.26.

Lynn A. Karoly and James H. Bigelow, The Economics of Investing in Universal Preschool
Education in California, (Santa Monica: RAND Inc., 2005).

David Elkind, Miseducation: Preschoolers at Risk (New York: Knopf, 1997): p.69.

Ibid., p.4.

Susanna Loeb, et al, The Influence of Preschool Centers on Children’s Development
Nationwide—How Much is Too Much? Policy Analyses for California Education, November
2005. http://pace.berkeley.edu/pace_stanford_berkeley.html

Michael Baker, Jonathan Gruber, and Kevin Milligan, What Can We Learn from Quebec’s
Universal Childcare Program?, (C.D. Howe Institute, February 2006).
http://www.cdhowe.org/pdf/ebrief_25_english.pdf

For a summary of current trends in universal programs, see Susan Block, “In the Beginning:
Good Preschools Have Much to Offer, But Should They Be Universal? Some Research
Suggests Otherwise,” American School Board Journal, November 2003, available at
www.asbj.com/2003/11/1103research.html (accessed January 18, 2005).

The National Education Goals Panel (NEGP) has been dissolved pursuant to congressional
mandate. NEGP publications are available online at govinfo.library.unt.edu/negp/.

For discussion on readiness and kindergarten curriculum, see Richard Rothstein, “Lessons: In
the Kingergartens, A Misguided Push,” The New York Times, March 21, 2001; Lorrie A.
Shepard, “Children Not Ready to Learn? The Invalidity of School Readiness Testing,”
Psychology in the Schools 34, no. 2 (1997): pp.85–97; Patricia L. de Cos, “Readiness for
Kindergarten: What Does It Mean?” California Research Bureau, California State Library,
CRB–97–014, December 1997; and Yolanda Bellisimo, Colin H. Sacks, and John R. Mer,
Deborah May, et al., “School Readiness: An Obstacle to Intervention and Inclusion,” Journal
of Early Intervention 18, no. 3 (1994): pp.290–301;

Other measures include physical health indicators, which are not addressed in this analysis. See
also Pascal Forgione, U.S. Commissioner of Education Statistics, “Early Childhood Education:
Critical Data Needs for a Critical Period of Child Development,” Testimony before the
Committee on Labor and Human Resources Hearing, U.S. Senate, December 17, 1999.

U.S. Department of Educations, National Center for Education Statistics, Digest of Education
Statistics, 2003, Table 43, available at nces.ed.gov/programs/digest/d03/tables/dt043.asp
(accessed January 18, 2005).

Reason Foundation

Kathryn Chandler, “Home Literacy Activities and Signs of Children’s Emerging Literacy,
1993 and 1999,” National Center for Education Statistics, Statistics in Brief, NCES 2000–026,
November 1999.

Nicholas Zill and Jerry West, “Entering Kindergarten: A Portrait of American Children When
They Begin School,” U.S. Department of Education, National Center for Education Statistics,
NCES 2001–035, January 2001: p.6.

U.S. Department of Education, National Center for Education Statistics, “America’s
Kindergartners,” NCES 2000–070, February 2000: p.16.

Ibid., p.17.

Ninety-six percent rank physical health as very important or essential, followed by 84 percent
ranking communicating needs as very important or essential, followed by 76 percent ranking
enthusiasm in approaching new activities as very important or essential. We do not examine
the second factor, communication, because there is no corresponding measure for this factor in
the NCES data. U.S. Department of Education, National Center for Education Statistics, Digest
of Education Statistics 1996, Table 48, available at
nces.ed.gov/programs/digest/d96/D96T048.asp (accessed January 18, 2005).

Parents report that 92 percent of children are eager to learn. However, teachers report that 75
percent are eager to learn—highlighting the subjective nature of these measures. U.S.
Department of Education, “America’s Kindergartners,” pp.39, 45.

U.S. Department of Education, Digest of Education Statistics 1996, Table 48.

Monica Mendoza, “Governor will push early-ed programs,” The Arizona Republic, January 11,
2004.

Monica Mendoza, “Kindergarten for all children, all day?; Discussion gaining wider national
audience,” Arizona Republic, December 12, 2004; Paul Davenport, “Controversial education
issues await legislators,” Associate Press State & Local Wire, December 6, 2004; and Le
Templar, “All-day kindergarten touted,” East Valley Tribune, February 17, 2004.

Janet Napolitano, testimony on Senate Bill 1362, February 16, 2004.

Monica Mendoza, “Seeking a fix for education Arizona looks to North Carolina for academic
model,” Arizona Republic, September 17, 2003.

Donald T. Campbell and Julian C. Campbell, Experimental and Quasi-Experimental Designs
for Research (Boston: Houghton Mifflin Company, 1963).

Alhambra School District No. 68, “Alhambra Elementary School District Comparison Study
Showing the Benefits of Preschool/Full-Day Kindergarten Education,” September 2003.

Ibid.

For background, see Anne T. Henderson and Nancy Berla, eds., “A New Generation of
Evidence: The Family is Critical to Student Achievement,” Center for Law and Education,
Washington, D.C.; and U.S. Department of Education, “Strong Families, Strong Schools,” U.S.
Government Printing Office, 1995–392–051, September 1994.

Chino Valley Unified School District No. 51, Office of the Superintendent, “Academic
Benefits of Full Day Kindergarten,” no date. This two-page report from the Office of the
Superintendent must be taken at face value because it does not include any supporting data.

Benchmark is defined as “on track to read at grade level.”

PRE-K EDUCATION

Tom Horne, “Kindergarten Report: Half Day and Full Day Programs,” Department of
Education, State of Arizona, Fall 2004.

For a comprehensive review of kindergarten research prior to the NCES findings, see James
Elicker, “Full-Day Kindergarten: Exploring the Research,” From Inquiry to Practice, Phi Delta
Kappa International, Bloomington, Indiana, 2000. Elicker concludes that “students
participating in full-day kindergarten consistently progress further academically during the
kindergarten year, as assessed by achievement tests, than students in either half-day or
alternate-day programs…There is not current, strong evidence that the academic achievement
gains of full-day kindergarten persist beyond first grade for all students….More research is
needed to examine the long-term impact of full-day kindergarten,” Ibid., pp.8–9.

Jerry West, Kristin Denton, and Lizabeth M. Reaney, “The Kindergarten Year: Findings from
the Early Childhood Longitudinal Study, Kindergarten Class of 1998–99,” U.S. Department of
Education, National Center for Education Statistics, p.11, available at
nces.ed.gov/pubs2001/2001023.pdf (accessed January 18, 2005).

Kristin Denton, Jerry West, and Jill Walston, “Reading—Young Children’s Achievement and
Classroom Experiences,” U.S. Department of Education, National Center for Education
Statistics, NCES 2003–070, July 2003, p.13, available at nces.ed.gov/pubs2003/2003070.pdf
(accessed January 18, 2005).

West, Denton, and Reaney, “The Kindergarten Year” Table 2, p.44, available at
nces.ed.gov/pubs2001/2001023.pdf (accessed January 18, 2005).

Amy Rathburn, Jerry West, and Elvira Germino-Hausken, “From Kindergarten Through Third
Grade: Children’s Beginning School Experiences,” U.S Department of Education, National
Center for Education Statistics, NCES 2004–007, August 2004, p.33, available at
nces.ed.gov/pubs2004/2004007.pdf (accessed January 18, 2005).

Ibid., p.24.

U.S. Department of Education, Digest of Education Statistics, 2003.

Andrew Coulson, Market Education (London: Transaction Publishers, 1999): p.179.

Between1959–60 and 1989–90, inflation-adjusted spending more than tripled. See Andrew
Coulson, Market Education, 32. From 1959–60 to 2001–02, average annual salaries for
elementary and secondary teachers rose from $36,000 to $44,000 (in constant dollars). See
U.S. Department of Education, National Center for Education Statistics, Digest of Education
Statistics, 2002, available at nces.ed.gov/programs/digest/d02/dt077.asp (accessed January 18,
2005). Pupil-to-teacher ratios in public schools have declined from one teacher for every 26
students in 1960 to one teacher for every 16 students in 2002. See U.S. Department of
Education, Digest of Education Statistics, 2003.

Sandra Feldman, Remarks to the AFT QuEST Conference, July 12, 2001, available at
www.aft.org/presscenter/speeches-columns/speeches/feldman071201.htm (accessed January
18, 2005). See also Ann Dryden Witte and Marisol Trowbridge, “The Structure of Early Care
and Education in the United States: Historical Evolution and International Comparisons,”
NBER Working Paper No. 10931, November 2004.

U.S. Department of Education, “Comparative Indicators of Education in the United States and
Other G-8 Countries: 2002.”

Reason Foundation

U.S. Department of Education, National Center for Education Statistics, “International
Comparisons in Fourth-Grade Reading Literacy: Findings from the Progress in International
Reading Literacy Study (PIRLS) of 2001,” NCES 2003-073, April 2003, p.5.

More than 50 percent of three-year-olds and 75 percent of four-year-olds in Germany, Italy,
and the United Kingdom attend center-based institutions. Ibid.

U.S. Department of Education, National Center for Education Statistics, “Elementary and
Secondary Education: An International Perspective,” NCES 2000–033, March 2000, p.56.

U.S. students outperformed 23 out of 26 participating countries. Students in Italy, Germany,
and France did not participate in this exam. See “TIMSS Highlights From the Primary
Grades,” available at timss.bc.edu/timss1995i/TIMSSPDF/P1HiLite.pdf (accessed January 18,
2005).

U.S. students performed better than those in the United Kingdom. Students in Italy, Germany,
and France did not participate in this exam. Ibid.

David J. Hoff, “U.S. Students’ Scores Drop by 8th Grade,” Education Week, December 13,
2000.

For a summary of math and science scores, see International Association for the Evaluation of
Educational Achievement, “Mathematics and Science Achievement in the Final Year of
Secondary School: IEA’s Third International Mathematics and Science Study,” February 1998,
available at isc.bc.edu/timss1995i/MathScienceC.html (accessed January 18, 2005).

U.S. students scored 471—significantly below the international average of 500. Ibid.

U.S. Department of Education, National Center for Education Statistics, “Elementary and
Secondary Education: An International Perspective,” NCES 2000–033, March 2000, p.51.

Reynolds et al., p.8.

Lawrence J. Schweinhart, “The High/Scope Perry Preschool Study Through Age 40,”
High/Scope Educational Research Foundation. See www.highscope.org/welcome.asp for
official publication date.

Zigler, Taussig, and Black, “Early Childhood Intervention” p.1000.

For a complete program description, see Lawrence Schweinhart and David Weikart, “The
Effects of the Perry Preschool Program on Youths through Age 15: A Summary,” in the
Consortium for Longitudinal Studies, As the Twig Is Bent–Lasting Effects of Preschool
Programs (Hillsdale, N.J.: Lawrence Erlbaum Associates, 1983): 71–81.

Lawrence J. Schweinhart, “Lasting Benefits of Preschool Programs,” ERIC Digest EDO–PS–
94–2 (January 1994).

Lawrence J. Schweinhart, Significant Benefits: The High/Scope Perry Preschool Study through
Age 27 (Ypsilante, Mich.: High/Scope Press, 1993): p.55; and Schweinhart, “Lasting Benefits
of Preschool Programs,” p.2.

Zigler, Taussig, and Black, “Early Childhood Intervention,” p.1002.

Ibid., p.1000.

Charles Locurto, “Beyond IQ in Preschool Programs?” Intelligence 15 (1991): pp. 299–305.

Ibid., pp.303–304.

PRE-K EDUCATION

Edward F. Zigler, “Formal Schooling for Four-Year-Olds? No,” in Early Schooling: The
National Debate, Sharon L. Kagan and Edward F. Zigler eds. (New Haven, Conn.: Yale
University Press. 1987), pp.30–31.

Ron Haskins, “Beyond Metaphor: The Efficacy of Early Childhood Education,” American
Psychologist, 44, no. 2 (February 1989): p.279.

Schweinhart, Significant Benefits, pp.59, 86, 106.

Deanna S. Gomby et al., “Long-Term Outcomes of Early Childhood Programs: Analysis and
Recommendations,” in The Future of Children, 5, no. 3 (Winter 1995): p.14.

Karoly and Bigelow, The Economics of Investing in Universal Preschool Education in
California.

Arthur J. Reynolds et al, Long-term Effects of an Early Childhood Intervention on Educational
Achievement and Juvenile Arrest, Journal of American Medical Association. Vol. 285, No. 18,
May 9, 2001.

Arthur J. Reynolds and Judy A. Temple, “Early Childhood Educational Intervention and LongTerm Developmental Outcomes,” Letters, The Journal of American Medical Association, Vol.
286, No. 15, October 17, 2001.

Mathew D. Thompson, “Early Childhood Educational Intervention and Long-Term
Developmental Outcomes,” Letters, The Journal of American Medical Association, Vol. 286,
No. 15, October 17, 2001.

http://www.stopreiner.org/facts.html

For background on the Abecedarian Project, see Frances A. Campbell and Craig T. Ramey,
“Cognitive and School Outcomes for High-Risk African-American Students at Middle
Adolescence: Positive Effects of Early Intervention,” American Educational Research Journal
32, no. 4 (Winter 1995): pp.743–72; and Craig T. Ramey, Frances A. Campbell, and Clancy
Blair, “Enhancing the Life Course for High-Risk Children: Results from the Abecedarian
Project,” in Jonathan Crane, ed., Social Programs That Work (New York: Russell Sage
Foundation: 1998): pp.179–180.

Campbell and Ramey, “Cognitive and School Outcomes for High-Risk African American
Students,” pp.750–752.

Campbell et al., “The Development of Cognitive and Academic Abilities: Growth Curves from
an Early Childhood Educational Experiment,” Developmental Psychology (March 2001):
pp.231–242.

Campbell and Ramey, “Cognitive and School Outcomes for High-Risk African American
Students,” p.766.

100

Bryant and Maxwell, “The State of Early Childhood Intervention,” p.34.

101

Spitz left Johnstone in 1989. Since then he has published several papers on the Abecedarian
Project, including Herman H. Spitz, “Some Questions about the Results of the Abecedarian
Early Intervention Project Cited by the APA Task Force on Intelligence,” American
Psychologist 52, no. 1 (January 1997): p.72; Herman H. Spitz, “When Prophecy Fails: On
Ramey’s Response to Spitz’s Critique of the Abecedarian Project,” Intelligence 17 (1993):
pp.17–23; Herman H. Spitz, “Spitz’s Reply to Ramey’s Response to Spitz’s First Reply to
Ramey’s First Response to Spitz’s Critique of the Abecedarian Project,” Intelligence 17

Reason Foundation

(1993): pp.31–35; Herman H. Spitz, “Early Educational Intervention Research and Cronbach’s
Two Disciplines of Scientific Psychology,” Intelligence 17 (1993): pp.251–255; Herman H.
Spitz, “Does the Carolina Abecedarian Early Intervention Project Prevent Sociocultural Mental
Retardation?” Intelligence 16 (1992): pp.225–237; and Herman H. Spitz, “Commentary on
Locurto’s ‘Beyond IQ in Preschool Programs?’” Intelligence 15(1991), pp.327–333.
102

Spitz, “Does the Carolina Abecedarian Early Intervention Project Prevent Sociocultural Mental
Retardation?”: pp.228–229.

103

Spitz, “Some Questions about the Results of the Abecedarian Early Intervention Project Cited
by the APA Task Force on Intelligence,” p.72.

104

Ibid.

105

Spitz, “Spitz’s Reply to Ramey’s Response to Spitz’s First Reply to Ramey’s First Response to
Spitz’s Critique of the Abecedarian Project,” p.35.

106

For example, Spitz explains that mothers of the experimental group were present at the testing
and assisted in the administration of some tests. That means that the mothers may have
provided their children with practice on some of the test items. Spitz, “Does the Carolina
Abecedarian Early Intervention Project Prevent Sociocultural Mental Retardation?”: pp.231–
232.

107

Campbell et al., “The Development of Cognitive and Academic Abilities: Growth Curves from
an Early Childhood Educational Experiment,” pp.231–242.

108

See U.S. Department of Health, Education, and Welfare, Office of Child Development,
“Recommendations for a Head Start Program by a Panel of Experts,” February 19, 1965: pp.1–
7. This report is now available from the Department of Health and Human Services,
Administration for Children, Youth and Families; and Lyndon B. Johnson, Public Papers of
the Presidents of the United States: Lyndon B. Johnson, Book 1, January 1 to May 31, 1965
(Washington: Government Printing Office, 1966): p.556.

109

U.S. General Accounting Office, “Head Start: Research Provides Little Information on Impact
of Current Program,” GAO/HEHS–97–59, April 1997.

110

Selection criteria included whether Head Start participation had occurred in 1976 or later,
whether studies compared outcomes for participants with those for children not attending any
preschool or another preschool, whether studies compared Head Start outcomes with test
norms, and whether studies used tests of statistical significance. Ibid., 10, pp.24–26.

111

Ibid., p.8.

112

Reproduced in ibid., p.48.

113

The synthesis study is McKey et al., conducted under contract for the Department of Health
and Human Services. Ruth McKey et al., “The Impact of Head Start on Children, Families, and
Communities,” U.S. Department of Health and Human Services, HHS 85–31193, June 1985.

114

Ibid., p.1.

115

Ibid., p.8, III–11.

116

Ibid., pp.12–13

117

Nicholas Zill et al., “Head Start FACES 2000: A Whole-Child Perspective on Program
Performance,” U.S. Department of Health and Human Services, HHS–105–96–1912, May
2003.

PRE-K EDUCATION

118

Laura Henderson, Kathleen Basile, and Gary Henry, “Prekindergarten Longitudinal Study
1997–1998 School Year Annual Report,” Georgia State University Applied Research Center
School of Policy Studies, April 1999, p.39.

119

Linda Schrenko, cited in James Salzer, “School Readiness the Same for Tots; Results
Unchanged Despite Pre-K,” The Florida Times-Union, November 1, 1999.

120

Gary T. Henry et al., “Georgia Pre-K Longitudinal Study: Final Report 1996–2001,” Andrew
Young School of Policy Studies, Georgia State University, May 2003: p.42, available at
aysps.gsu.edu/publications/GPKLSFinalReportMay2003.pdf (accessed January 18, 2005).

121

Ibid., p.42; and Linda C. Schrenko, State School Superintendent, “Superintendent Schrenko
Releases State NAEP Mathematics and Stanford 9 Scores,” press release, August 2, 2001,
available at www.doe.k12.ga.us/_documents/curriculum/testing/scores_stanford_01.pdf
(accessed January 18, 2005).

122

“Pre-K Report: Local Flexibility Works,” Andrew Young School of Policy Studies news
release, May 19, 2003, available at aysps.gsu.edu/news/release/prekreport.htm (accessed
January 18, 2005).

123

Ibid.

124

Gary Henry, cited in ibid.

125

Duane D. Stanford, “Pre-k’s true story,” Atlanta Journal-Constitution, February 23, 2000.

126

Kavan Peterson, State of Education: Who Makes the Grade?, January 26, 2006, Stateline.org,
http://www.stateline.org/live/ViewPage.action?siteNodeId=136&languageId=1&contentId=83
484

127

Michael Kelley, Joseph Tobin, and Karen Ortiz, “The Condition of Early Childhood Education
and Care in Arizona: 2004,” Arizona State University Education Policy Studies Laboratory,
May 28, 2004: executive summary and p.2.20.

128

John Boehner, Taking Money from Children: Financial Abuse and Mismanagement in the
Head Start Early Childhood Program, U.S. House Committee on Education and the Workforce,
March 18, 2005.

129

“Preschool Problems; Rocky Road for Promising Program,” The Record (Bergen County, NJ)
March 28, 2005.

130

Jean Rimbach, “Gaps Found in Preschool Quality,” The Record (Bergen County, N.J.), March
25, 2004.

131

Troy Anderson, “Fraud Eats Funds for Child Care: D.A.’s Deputies Report Millions Looted,”
Los Angeles Daily News, January 31, 2006. http://www.dailynews.com/news/ci_3459488

132

Rachana Rathi, “Audit Findings Jeopardize Head Start Funding for County Preschool
Centers,” p.B3, Los Angeles Times, Auust 5, 2005.

133

Kelley, Tobin, and Ortiz, “The Condition of Early Childhood Education,” pp.2.15-2.16

134

Ibid., pp.2.15, 2.18.

135

For instance, see Coulson, Market Education; and Milton Friedman, “Public Schools: Make
Them Private,” Cato Institute Briefing Paper no. 23, June 23, 1995, available at
www.cato.org/pubs/briefs/bp-023.html (accessed January 18, 2005).

Reason Foundation

136

The diversity of programs and multiplicity of funding streams makes it difficult to quantify
total expenditures. We believe the best available estimates are documented in Kelley, Tobin,
and Ortiz, “The Condition of Early Childhood Education and Care in Arizona: 2004”: p.2.12.
Kelly, Tobin, and Ortiz report $12 million for Early Head Start; $87 million for Head Start
(excluding tribal and migrant worker programs); $19.5 million for the Early Childhood Block
Grant; $9.95 million from block grant funds for preschool; $148.7 million in childcare
subsidies; $134.2 million on kindergarten (excluding multiple funding streams utilized by
school districts); and $171,335 for kith-and-kin programs—for a total of $411.5 million.

137

Ibid., p.2.12.

138

To derive this estimate, we divided the estimated $134 million spent by the state on
kindergarten by the number of kindergartners in public and charter school kindergartens—
68,748. See ibid., p.2.10, 2.12.

139

An estimated 42 percent of families with children ages 0–5 have incomes equal to or below
185 percent of the federal poverty line. For this calculation, we assumed 42 percent of
kindergartners would qualify for the grant (28,874 children). See Charles Bruner et al., “Early
Learning Left Out: An Examination of Public Investments in Education and Development by
Child Age,” Voices for America’s Children and the Child and Family Policy Center, February
2004: p.24, available at
www.voicesforamericaschildren.org/Content/ContentGroups/PublicationsVoices/ECE1/ELLO/ELLO.pdf (accessed January 18, 2005).

140

The $265 million includes $87 million for Head Start, $19.5 million from the Early Childhood
Block Grant, $9.95 million from block grants used for pre-school and $148.7 in child care
subsidies. We exclude Early Head Start money, which is normally directed at much younger
children. There are an estimated 76,560 four-year-olds in Arizona. See Kelley, Tobin and
Ortiz, “The Condition of Early Childhood Education and Care in Arizona: 2004,” pp.2.7, 2.12.

141

An estimated 42 percent of families with children through age 5 have incomes equal to or
below 185 percent of the federal poverty line. For this calculation, we assume 42 percent of
four-year-olds would qualify for the grant (32,155 children). See Bruner et al., “Early Learning
Left Out,” p.24.

142

Vicki Murray and Ross Groen, “Survey of Arizona Private Schools: Tuition, Testing, and
Curricula,” Goldwater Institute Policy Report no. 199 (January 5, 2005), available at
www.goldwaterinstitute.org/article.php/506.html (accessed January 18, 2005).

143

U.S. Department of Education, Digest of Education Statistics, 2002, U.S. Department of
Education, Table 61.

144

Arizona School Readiness Task Force, “Growing Arizona,” Children’s Action Alliance, July
2002.

145

National Institute for Early Education Research, “Cost of Providing Quality Preschool
Education to America’s three- and four-year-olds,” Fast Facts, available at
nieer.org/resources/facts/index.php?FastFactID=5 (accessed January 18, 2005).

146

Shikha Dalmia and Lisa Snell, Universal Preschool is Inviting Universal Disaster, San
Francisco Chronicle, December 4, 2005.

147

Norma Greenaway, “More than Half of Canadian Preschoolers in Child Care,” The Calgary
Herald, April 6, 2006.

148

Le Templar, “All-day kindergarten touted.”

PRE-K EDUCATION

149

S. V. Date, “Lawmakers urge vouchers for pre-k: They say vouchers are key to universal early
education,” Palm Beach Post, March 22, 2004.

150

Brent Kallestad, “Governor Bush signs pre-k bill,” Associated Press, January 3, 2005.

151

Alex Penelas, cited in S. V. Date, “Lawmakers urge vouchers for pre-K.”

152

Linda Carlton, cited in Mary Ellen Klas, “Pre-K Vouchers Program Provokes Dispute,” Miami
Herald, April 5, 2004, available at www.floridians.org/newsf/04/040504.html (accessed
January 18, 2005).

153

“Statewide Survey of Parents with Children in VPK Reveals a High Level of Satisfaction with
the Program, Office of Early Learning Program Evaluation, January 2006.
http://www.floridajobs.org/earlylearning/Perf_Account/Parent%20Report.doc

154

Anjali Athavaley, “Private Preschools are Expanding in South Florida and Nationwide, The
Miami Herald, December 28, 2005 p. C1

155

For sample tax credit legislation see: Lori Drummer, “Model Scholarship Legislation for
Preschool, Pre-K Adopted by Alec,” School Reform News, July 1, 2005.

156

Ibid.

157

“More Money to Rescue Playgroups,” BBC News, March 29, 2000.

158

Ibid. See also “Playgroup Places Continue to Fall,” BBC News, October 1, 1999; and “Call for
Shake-up in Preschool Admissions,” BBC News, October 15, 1999.

159

Gary T. Henderson and Craig S. Gordon, “Competition in the sandbox: A Test of the effects of
preschool competition on educational outcomes,” Journal of Policy Analysis and Management,
V. 25, 2006, pp.97–127.

160

Janet Napolitano, Remarks to Supporters of the National Task Force on Public Education.

161

To learn more about the U.S. experience with school choice, we recommend the research
summary available at
www.friedmanfoundation.org/schoolchoiceworks/schoolchoiceresearch.html (accessed January
18, 2005). See also www.friedmanfoundation.org/schoolchoice/index.html (accessed January
18, 2005).

162

Rathburn, West, and Germino-Hausken, “From Kindergarten Through Third Grade,” p.33.

163

Henderson, Basile, and Henry, “Prekindergarten Longitudinal Study 1997–1998 School Year
Annual Report,” p.39.

164

McKey et al., 8, pp.III–11.

Reason Public Policy Institute
3415 S. Sepulveda Blvd., Suite 400
Los Angeles, CA 90034
310/391-2245
310/391- 4395 (fax)
www.reason.org

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.6
Linearized                      : Yes
XMP Toolkit                     : 3.1-702
About                           : 0596c0a2-44d5-11d9-bcd6-a163838728e7
Create Date                     : 2006:05:10 10:42:57-07:00
Modify Date                     : 2006:05:10 10:45:50-07:00
Metadata Date                   : 2006:05:10 10:45:50-07:00
Creator Tool                    : Adobe InDesign CS2 (4.0.1)
Thumbnail Format                : JPEG
Thumbnail Width                 : 256
Thumbnail Height                : 256
Thumbnail Image                 : (Binary data 13648 bytes, use -b option to extract)
Version ID                      : 1
Instance ID                     : uuid:47070921-6130-49c9-a3cf-5da11be05555
Document ID                     : adobe:docid:indd:ba255128-dbc0-11da-a364-91e57ae33bb1
Rendition Class                 : proof:pdf
Derived From Instance ID        : ba255127-dbc0-11da-a364-91e57ae33bb1
Derived From Document ID        : adobe:docid:indd:0596c0a1-44d5-11d9-bcd6-a163838728e7
Derived From Version ID         : 1
Manifest Link Form              : ReferenceStream, ReferenceStream, ReferenceStream, ReferenceStream
Manifest Placed X Resolution    : 72.00, 72.00, 72.00, 72.00
Manifest Placed Y Resolution    : 72.00, 72.00, 72.00, 72.00
Manifest Placed Resolution Unit : Inches, Inches, Inches, Inches
Manifest Reference Document ID  : adobe:docid:photoshop:b5a72cb2-f3e8-11d6-a9ac-de677b19984c, adobe:docid:photoshop:66b432da-f3e7-11d6-af79-cdee9789b5bd, adobe:docid:photoshop:b5a72cb2-f3e8-11d6-a9ac-de677b19984c, adobe:docid:photoshop:66b432da-f3e7-11d6-af79-cdee9789b5bd
Format                          : application/pdf
Producer                        : Adobe PDF Library 7.0
Trapped                         : False
Has XFA                         : No
Page Count                      : 56
Creator                         : Adobe InDesign CS2 (4.0.1)

EXIF Metadata provided by EXIF.tools

CD 824 B7abd1fc30bdf33cd824db3b102c4db0

Navigation menu

Versions of this User Manual:

Views

Navigation