.115!

hrs.)- .5:

3.3....

. a... .

m a: :55...
h

5 I. r
.1... 8‘ ..
f‘otpfz

... [5
4.5.2....“
.1 .

4

 

:.1.C.....a.l!. )gi
. A 2: .- . .
. .5; I

.!

 

This is to certify that the
dissertation entitled

EARLY CHILDHOOD EDUCATION: ORIGINS, THEORIES
AND POLICY REALITIES

presented by
Rachel Fulcher Dawson

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Education Policy

,6 Cut/Va Jaime
Major Professor’s Signature
I 2/ 3 )0 8

 

 

Date

MSU is an Affirmative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DATE DUE DATE DUE DATE DUE

 

 

EARLY CHILDHOOD EDUCATION: ORIGINS,
THEORIES AND POLICY REALITIES

By

Rachel Fulcher Dawson

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Education Policy

2008

ABSTRACT

EARLY CHILDHOOD EDUCATION: ORIGINS,
THEORIES AND POLICY REALITIES

By

Rachel Fulcher Dawson

This dissertation explores Early Childhood Education (ECE) Policy fiom both a
theoretical and empirical perspective drawing on sociology and political science. I first
explain the size, scope and status of this policy issue. This sets the stage for the
subsequent analysis and demonstrates the trends and the conflicting data in this policy
area. Then I turn to three theories of policymaking in order to propose how ECE policy
might better be understood, evaluated and enacted. Using these theories - multiple
streams, issue framing and diffusion — I present three different analysis of ECE policy.
First I focus on problem definitions by viewing the history of ECE policy in the US.
through Kingdon’s policy streams lens. I break ECE policy into 3 general time periods-
pre-Head Start era, Head Start era, and post-Head Start era. I find clear evidence of
considerable differences in the theoretical and operational problem definitions of ECE
during each era. Next I present results of a case study of framing in two different states
over a 25 year period. I find evidence that the framing of ECE has wavered between that
of education frames and welfare frames, though two frames that distinguish ECE from
these others are the “scientific findings” or “evidence” frame and the “economic
investment” frame. Finally, I present the results of a 50-state analysis of state adoption of
ECE policy. I find that percentage poverty, working women and a liberal government

ideology increase the log odds of a state adopting preschool policy in a given year. Most

significant is that the ratio of education to welfare spending (a proxy variable to these two
domains’ relative strength in a given state year) is significant in increasing these log
odds. I also find that early adopting states are most likely to “reinvent” or innovate in
this policy area and reinvention is most likely in the areas of teacher policy and funding.
Finally, I find that states not adopting ECE policy have significantly higher Head Start
enrollment supporting (though not significantly) the notion that federal policy created a
pressure valve effect in these states that found federal policy sufficient to meet their ECE

policy needs.

To Mom, Dad, Matthew, Joanna, Josh, Martha, Scott, Grace, Ellie and Caroline —

the best, most loving, most consistent and most patient teachers in my life.

iv

ACKNOWLEDGEMENTS

I have many people to thank for their help with this dissertation and with the
completion of my PhD.

I wrote this dissertation during the “early years” of three very important people in
my life, my daughters, Grace, Ellie and Caroline (who was born during the dissertation
year)! In theory this gave me a personal perspective on what it is that magically takes
place during transformation of a newborn into a toddler into a preschooler into a
kindergartener. For my kids, I see how each went through different phases and
developmental changes at their own unique pace — while still basically being raised in the
same environment. For me and my husband, I see how it has been magical and
miraculous to witness the beauty and brilliance of each of them — while coping with the
physical, mental and emotional challenges of parenting three young children. The
“personal perspective” then is simple: no matter what policy I have evaluated or
compared or analyzed, nothing has yet to capture for me what I have personally
experienced. It is as a researcher, then that I have tried to understand how the public role
has developed in the life of young children - how it is that as a society, it is absolutely in
the best interest of everyone that all children have the best possible early experience. The
challenge is in finding out how that public role meets the personal, private choices of
families, parents and children.

My own early experience was molded by my parents, Nancy and Tony Fulcher.
Through a combination of love, strength, sacrifice and faith, they saw to it that me and

my four siblings had access to the very best education has to offer: development of each

individual learner, challenges in and out of the classroom, outstanding teaching, high
expectations met with support and resources. There is no way to properly acknowledge
what my parents - my first teachers - have meant for me. Everything would not be
overshooting - not by a mile.

My siblings, Matthew, Joanna, Josh and Martha were and are my constant
companions, rivals and friends. Having them in my life is a constant reminder that my
experience and interpretation therein are not so unique, and that in statistical terms, an “N
of 1” is informative, but rarely captures the full effect of anything!

My husband Scott has been my constant supporter and champion for the past 10
years and especially during this dissertation process. His love, humor and friendship
have been life-support for me when I doubted myself, my path and my stamina. I think
he did best by encouraging me without ego of his own in a process — a dissertation— that I
think requires just enough ego of its author to actually complete it and more than enough
humility to realize it’s not a masterpiece!

I wanted to use this space to acknowledge each and every teacher I have had in
my life. It is overwhelming and informative to actually sit down and think about this list
of people. I remember the name of every teacher for every grade from preK to 12 (34
people). Oddly enough, I had a much harder time remembering all the great professors I
had as an undergrad at Marquette (though I can estimate this number to be around 37)
and then as a grad student at Maryland (19). Here at MSU, I’ve had the great fortune of
being taught and working with Drs. Sedlak, Plank, Fairweather, Arsen, Sykes, Youngs,
Fairweather and Weiland in the College of Education and Drs. Jacoby, Schneider, Creel-

Davis, Hammond and Kautz in the Department of Political Science. I truly believe and

vi

know this dissertation to be a cumulative product of these 103 people — their patience,
inspiration, encouragement and belief in me as a student gave me a lifelong love of
learning and a desire to always, always search for knowledge and truth in a helpfiil way.

I come to the field of education policy from both a teacher’s and a political
perspective. Though I taught almost a decade ago, I still recall often the amazing
experience of working, learning and teaching at Notre Dame Middle School in
Milwaukee, Wisconsin. I am humbled by teachers who work their entire careers in the
classroom —— dedicated to children and to learning in a way that I know I cannot be
directly. I must thank the Rev. Timothy O’Brien, Ph.lD. for grounding my political
perspectives and belief in the American governmental system. And the only two
politicians I have ever worked for — Rep. Timothy Roemer (IN) and Sen. Herb Kohl (WI)
— make me proud of America and of what it means to be political: no scandals, no
nonsense, hard work, strong ethics and morals, fieedom-loving.

Finally, the committee of five people who helped move me from ABD to PhD is
quite extraordinary — Drs. Gary Sykes, Barbara Schneider, Peter Youngs, Sandra
Schneider and Larry Schweinhart. They are all obviously amazing scholars and
policymakers in their own fields, and they let me utilize their expertise and experience to
further my own work. In addition, they were extremely supportive and patient on a

personal level as my dissertation year proved to be quite tough for a variety of reasons.

vii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. viii
LIST OF FIGURES .................................................................................. x
INTRODUCTION .................................................................................... 1
CHAPTER 1: Current and Historical Status of ECE .......................................... 13
Current Status ............................................................................... 13
Historical Trends ........................................................................... 20
Areas of Consensus ........................................................................ 26
Countertrends ............................................................................... 32
Major countertrends: teacher quality and short-term evidence .............. 34
Conclusions ................................................................................. 36
CHAPTER 2: Understanding ECE Policy — A Theoretical Proposition ..................... 38
The Policymaking Process: Frames, Streams and Windows ......................... 39
Multiple Streams, the Garbage Can Model and Kingdon .................... 40

Issue Framing Theory ............................................................ 43

Policy over Time: Diffusion and Innovation ........................................... 48
Theory and ECE Policy ................................................................... 52
CHAPTER 3: Problem Definitions in ECE Policy ............................................. 54
Pre-Head Start Policy Era ................................................................. 56
Problem Definitions in the Pre-Head Start Era ................................ 57

Children and Teachers in Pre-Head Start Era ................................. 64

Government Role in ECE in Pre-Head Start Era .............................. 65

Conclusions on the Pre-Head Start Era ........................................ 68

Head Start Policy Era ..................................................................... 70
Problem Definitions in the Head Start Era .................................... 72

Race ........................................................................ 73

Poverty ..................................................................... 75

Educational Outcomes ................................................... 78

Problem Definitions and Head Start ................................... 80

Children and Teachers in the Head Start Era .................................. 88

Government Role in Head Start Era ............................................ 90

Conclusions on the Head Start Era ............................................. 92

Post Head Start Era Policy ................................................................ 94
Problem Definitions in the Post-Head Start Era .............................. 98

Children and Teachers in the Post-Head Start Era ........................... 103

Government Role in Post-Head Start Era .................................... 104

Problem Definition in ECE Policy ...................................................... 105

viii

CHAPTER 4: Framing Early Childhood Education Policy ................................. 108

Issue Framing and Frame Analysis... .................................................. 109
Methodology and Findings .............................................................. 111
Oklahoma: A study of universal preschool enactment .............................. 112
Michigan: A study of mixed state involvement ....................................... 117
Analysis and Conclusions ............................................................... 122
CHAPTER 5: A SO-State Diffusion and Innovation Study of ECE Policy ............... 123.
State Preschool as a Policy Innovation ................................................ 124

Data and Methodology ................................................................... 126

Policy Domain Strength Variable ............................................. 129

Diffusion Tests ................................................................... 129

Political Variables ............................................................... 130

Social and Economic Variables ................................................ 131

School Variables ................................................................. 132
EHA Results ............................................................................... 132
Reinvention of Preschool and ECE Policy in the States ............................. 134
States without Preschool Policy ......................................................... 136
Conclusions ................................................................................ l 38
APPENDICES ..................................................................................... 140
REFERENCES .................................................................................... 150

ix

LIST OF TABLES

Table 1.1: Care Arrangements for Children under 6 ........................................... 14
Table 1.2: Estimates of children served by major public programs .......................... 19
Table 2.1: Comparison of Three Policy Theories and Proposed Application to ECE
Pohcy .............................................................................. 141

Table 4.1: Oklahoma Governors, by Term, Political Party and ECE Policy ............... 113
Table 4.2: Oklahoma Governors by Term and Divided or Unified Government Status.113
Table 4.3: Oklahoma: Frames used in Education, Welfare and ECE ....................... 115
Table 4.4: Oklahoma Education Frames ........................................................ 1 16
Table 4.5: Oklahoma Welfare Frames .......................................................... 116
Table 4.6: Oklahoma Early Childhood Frames ................................................ 116
Table 4.7: Michigan Governors by Term, Political Party and ECE Policy ................ 118
Table 4.8: Michigan Governors by Unified or Divided Government Status ............... 118
Table 4.9: Michigan: Frames used in Education, Welfare and ECE Policy ............... 120
Table 4.10: Michigan Education Frames ....................................................... 121
Table 4.11: Michigan Welfare Frames ......................................................... 121
Table 4.12 Michigan Early Childhood Frames ................................................ 121
Table 4.13: Policy Controversies, Framing Processes and Political Opportunity. . . . . 143
Table 4.14: Issue Framing Definitions and Classifications ................................... 144
Table 5.1: State Adoption of Preschool Policy, 1978-2000 .................................. 148
Table 5.2: State Adoptions of Preschool Policy (Hazard Ratios), 1978-2000 ............. 149
Table 5.3: ECE Policy Reinvention in the States: 1965 to present .......................... 135
Table 5.4: Types of Reinvention by States ..................................................... 136

Table 5.5: Head Start and Special Education Percentage Enrollments by
Non-Adopting States ..................................................................... 137

xi

Figure 1.1:
Figure 1.2:
Figure 1.3:
Figure 1.4:
Figure 1.5:
Figure 1.6:
Figure 2.1:
Figure 2.2:
Figure 2.3:
Figure 3.1:
Figure 3.2:
Figure 4.1:
Figure 4.2:
Figure 4.3:
Figure 4.4:

Figure 4.5:

LIST OF FIGURES

Working Women by presence and age of youngest child: 1975-2004 ......... 21
Historic Enrollment in Preschool and Kindergarten: 1965-2003 ............... 22
Head Start Spending .................................................................. 23
Head Start Children Served .......................................................... 23
Adjusted Head Start Spending per Child .......................................... 24
Number of States with State Preschool Policy1978-2001 ........................ 26
MS Theory and Kingdon ............................................................. 42
Framing and Problem Definition .................................................... 44
Diffusion Curve Model ............................................................... 50
Three Eras of ECE Policy Timeline ................................................ 55
Problem Definitions for Pre-Head Start, Head Start and Modern Eras ....... 142

Oklahoma expenditures data for education, welfare and totals ................ 114
Michigan expenditures data for education, welfare and totals ................. 119
Timeline of 2-state data and Federal policies 1980-2001 ...................... 145
Oklahoma Frames for Education, Welfare and ECE ........................... 146
Michigan Frames for Education, Welfare and ECE ............................. 147

xii

Introduction

By much estimation, Early Childhood Education (ECE) is a hot policy topic.
Politicians, educationalists, economists and celebrities champion its importance for
children and for society. Most families send young children to some ECE program
before they begin kindergarten. All levels of government are involved in ECE provision,
regulation and administration. Non-governmental organizations and advocacy groups
champion the importance and need for it. Research is conclusive that quality programs
lead to dramatic benefits for children and society and that early childhood is absolutely
the most dynamic and critical growth period of an individual’s life. And all of these
factors of ECE policy are growing: more people as champions, more government
programs, and more research evidence. In this way, it seems that there is both momentum
and consensus around this policy issue with growth in programs and agreement on
funding and importance. Yet despite such seeming momentum and consensus, ECE
policy in the United States remains enigmatic with erratic policy growth across the
nation, conflicting views about the role of government in young children’s development,
inconsistent applications of scale-up from known successful programs and instable
funding and funding sources.

A quick snapshot shows momentum at most every level. First of all, most kids -
60 percent - under age five are in some sort of non-parental care and this is higher for
children ages three (65%) and four (79%) who are predominantly in formal care
arrangements such as childcare centers, preschools and nursery schools (Mulligan, et a1,
2005). The federal government has consistently funded Head Start regardless of partisan

control of the executive or legislative bodies, with increases at or above inflation since its

inception (Vinovskis, 2005; Head Start Bureau, 2006). All but twelve states have
adopted state-funded preschool programs, with over $2.8 billion state-dollars currently
spent on more than 800,000 kids (Barnett, et a1, 2006). This represents an increase in
funding of over 7.5 percent (adjusted for inflation) and over 100,000 kids in just the last
five years alone (2000-2005) (Barnett, et al, 2006). In addition, K-12 schools and
districts have a growing interest in ECE policy as 822,000 kids are now in preschools
housed in public schools and around 19,900 (35%) of all public elementary schools now
have prekindergarten programs on-site (N CES, 2003). Recent non-legislative efforts
provide additional levity to an ECE-push with the “adequacy” lawsuits in four states
including prekindergarten provision as part of the constitutional financial obligation of a
state (ACCESS, 2006). Recent efforts at statewide compliance with the No Child Left
Behind Act have led many states to align preschool standards with elementary ones-
forty-one states have now done this (Education Week, 2007).

Along with this momentum of activity and seeming growth in the ECE policy
area is a seeming consensus-everyone supports ECE generally speaking. Coalitions have
formed around such specific endeavors as Universal Prekindergarten (the PreK—Now
Group), Smart Start collaborations in the states, promoting NAEYC standards and
accreditation, large statewide initiatives and referenda (Florida and California most
recently). People from both political parties are supportive of ECE policy as are many
otherwise-polarized ideologues (for example, AEI, CATO, Manhattan Institute,
Children’s Defense Fund, Institute for Policy Studies, Brookings are think tanks across
the ideological spectrum and have all publicly recognized the importance of early

childhood). In addition, prominent economists —most notably the Nobel Laureate, James

Heckman — and well-known celebrities —most recently Rob “Meathead” Reiner, actor and
producer — have been able to draw national attention and bring in new audiences to the
ECE policy debates. Public opinion data demonstrate that an overwhelming majority of
people want increases in funding for preschool and other ECE initiatives-a recent NIEER
poll notes that 87 percent support state funding for preschool (N IEER, 2002). Evidence
from two different strands of research - neuroscience and random-assignment studies of
ECE programs — now converges along a common dimension: the experience and
environment of young children in early childhood has a direct impact on their immediate
learning and well-being as well as a significant long-term impact that affects each child
and society.

An examination of the seeming momentum and consensus in ECE policy is
particularly timely in light of current changes happening at the federal and state level of
government in K-12 education reform policy. In response to NCLB, states have
ratcheted up (or in some cases created from scratch) statewide standards and systems of
accountability in order to more closely align what is taught by teachers with what is
tested and what is learned. As the “feeder” system of graded schools, ECE programs and
schools are a natural locus of attention for K-12 reforms.

And yet, the momentum and consensus belie the challenges facing ECE policy
and policymakers. Countertrends demonstrate that while there is more attention and
more fimding than ever for ECE programs, the main ingredient from research in what
makes programs affective —quality- is often not prioritized in policymaking. By many
measures (teacher training, program quality) has plateaued and in some areas has gone

down over time — one need look no further than a program like the Child Care

Development Fund (CCDF) which funnels more than $5 billion per year to support child
care provision for low-income and poor families, but has minimal requirements for
providers, for program or for outcomes. Likewise, the seeming consensus masks the fact
that while many generally support ECE policy and laud the importance of early learning,
there are vastly different views of everything specific to ECE policy from state’s roles,
parental rights, age requirements, targeting vs. universal, and the like. In other words, the
consensus in ECE of a diverse group of supporters could be much like the “empty vessel”
charter policies in the states where a diverse set of supporters led to political compromise
resulting in widely different laws and implementation realities (Wells, et a1, 1999). A
prime example of such support is seen in the wide variety of policies at the federal level —
some twenty childcare and early learning programs- and the range of government
involvement at the state level (thirty-eight states have forty-eight preschool programs, not
to mention child care and other ECE programs). It quickly becomes clear that support
does not mean consensus policy or programs.

The amalgam of policies across the states alone makes it clear that momentum
and consensus are incomplete ways of understanding ECE policy today. It is not so
simple to say that because most kids are already in ECE programs, because most states
fund programs and because the federal government has been involved for so long, there is
clear forward movement in this issue area. Rather, a more thorough understanding of
how the existing policies have‘developed so differently is needed to shed light on what
factors come into play in developing ECE policy. There is an inherently unclear picture
of ECE policy in the US. as it encompasses welfare policy (aimed at poor families and

children), education policy (aimed at all children with learning outcomes) and family

policy (less governmental in nature, but aimed at families and their support systems
generally). This vagueness and variation is precisely what I aim to explore and hopefully
explain in a helpful way. I suggest that ECE policy is a reflection of a mixture of
competing political goals and a history of changing definitions and frames of
understanding.

Research Program. Methods and Findings

My research program centers on one line of questioning: What explains variation
in ECE policy? I answer this by first laying out the current data and historical trends for
ECE in the United States. Through a set of complementary theoretical lenses I then
explore what elements of the policymaking process have led to the existing status of ECE
policy. I use a mixture of methods: historical analysis of problem definition, qualitative
analysis of frames used in policymaking in two states, and quantitative analyses of 50-
state data.

I begin by presenting the current statistics relevant to this policy area as well as
trend data to situate it historically. What is evident from this data is that ECE has
certainly grown in usage, public spending and types of programs. While Head Start and
the Child Care Development Fund dominate federal policy in terms of dollars and
numbers of children served, they are vastly different programs with different goals,
regulatory controls and funding streams. In addition to these two large programs, there
are more than a dozen other federal programs that serve young children, many of which
serve the same populations as Head Start and the CCDF. Critical to an understanding of
ECE policy today is the development of a broad set of state programs and policies most

recently complied and rated by the National Institute of Early Education Research

(Barnett, et al, 2006). What is most remarkable in the current status of ECE is that it is
consistently funded at the federal level, it is increasingly funded by states, but that it
remains, in the aggregate, a largely private endeavor — families still bear on average
around 60 percent of the costs for ECE (Barnett and Masse, 2002). What is also
remarkable is that while there is great deal of consistent and uncontroversial evidence
from both the hard sciences (neuroscience) and the social sciences (education) of the
importance of early learning and early experiences, the programs and policies put in place
do not always require or fund the necessary ingredients of quality programs and quality
staff (Schweinhart and Fulcher-Dawson, 2006). A closer look reveals some evidence in
recent years of a leveling-off or plateauing of ECE demonstrating that access to care is no
longer a major issue. However, this plateauing masks the fact that there are still constant
trade-offs made in this policy area whereby more children are often brought into the fold
of public programs at the expense of quality concerns such as teacher training and class
size.

In order to understand the issue of ECE, it is necessary to consider a broader
perspective of how any policy changes and grows over time and across issue areas. As
such, I explore three theories that attend to the relevant considerations of ECE policy.
While these are not comprehensive theories of policymaking nor do I explore the entire
range of such theories, these theories are particularly chosen because they address such
things as time, social movements, and political “tipping points”, multiple issue streams
and innovation in the states. Such theories help inform any understanding of ECE policy
and outcomes by highlighting how issues evolve and change over time and in different

institutions. Furthermore, these theories help shed light on the apparent mismatch

identified previously between the seeming consensus and momentum surrounding ECE
policy and the widely variable and tenuous nature of such policy. These theories help
clarify how issues are defined, to what effect and what factors make enactment of policy
most likely.

Diffirsion and innovation theory focuses on the process through which
governments adopt new programs (Berry and Berry, 1999). The theory suggests that there
are variables that must be considered when examining why a government enacts a given
policy-in particular, state policy enactment depends on variables internal to the state as
well as variables external to it (such as other states’ adoptions, federal policy, national
economic trends). Kingdon’s theory of agenda setting and multiple streams suggests that
there are three streams: problems, policies and politics through which issues flow and at
certain points in time (usually accompanied by a policy entrepreneur) all three are
combined leading to attention on the agenda and action by policymakers (Kingdon,
1985). The idea is that policy growth and change is often incremental and predictable,
but that major changes can and often do happen demonstrating that policy change
depends both on the agenda and the issue definition. This ties into a third theory - issue
flaming - developed primarily in sociology, particularly, the theories of social
movements. This theory compliments the others as it suggests that the crux of social
movements is in how an issue is flamed, such that the flame focuses the public and the
debate on certain things and thus renders other facts and arguments moot (Schon and
Rein, 1992). Taken together, these three theories posit how an issue gets on the political
agenda, how it then gets enacted and how the definition and flaming of the issue moves

society in a given direction on this issue.

Using Kingdon’s theories of problem definition and policy streams, I examined
the history of ECE policy in the United States. I did this by separating this history into
three “eras” of ECE policy: the pre-Head Start era (roughly beginning in the 18203 with
infant schools and ending in the 19403 with Lanham Act child care centers); the Head
Start Era (from early 19503 through the late 19703); and the Modern or Post-Head Start
era (flom the early 19803 to present). I found that in the pre-Head Start era, there was
limited attention to early learning and that problem definitions wavered between the
developmental needs of children generally and the handicap of poverty in early
childhood. The “problem” for the two early twentieth century policies was simply
economic- out of work teachers in the Great Depression and later working mothers
during WWII who needed custodial care for their young children. The Head Start era
saw a rapid expansion of ECE policy most notably with the landmark Head Start
legislation and program in 1965. Yet the problems this policy set out to solve were
crippling as it was part of a larger agenda to end poverty and racial inequality. The
problems of poverty and racial inequality were large and deep rooted and as the idea of
an early childhood pro gram to mitigate racial and economic inequalities for young
children arose, so too did the political expediencies of making it a massive program at the
expense of major design flaws. As such the problem quickly also became the Head Start
policy itself as there was enormous pressure for this popular program to carry the weight
of the less popular War on Poverty programs on its preschool shoulders. Finally, the
Modern Era of ECE policy has seen an explosion in almost every policy direction as the
problem definitions multiplied and policy solutions were made at many levels of

government and more notably in private homes and families.

In a separate study, I looked at how ECE has been flamed as an issue by
policymakers. I looked at State of the State Addresses for the past twenty-five years for
two states — Michigan and Oklahoma to examine the flames used by these governors. I
compared ECE flames with those of welfare policy and education policy as these are the
two larger areas of domestic policy between or within which ECE is situated. What I
find is that flames used mirror those of both welfare and education, but ECE also enjoys
the “evidence” or “research” flame whereby ECE is flamed in terms of the evidence of
the long-term findings of Perry and Chicago preschools. In addition, the economic flame
was used more recently by governors in both states to present ECE as an “investment”
with sure returns for the state.

A final study involved using event history analysis (EHA) where I examined the
adoption of state ECE policies over a twenty year period. I tested for both the typical
internal and external variables of diffusion as well as for a policy stream variable that I
call “policy domain-strength” which posits that the relative strength of education to
welfare as an issue in a state increases the adoption of ECE policy. I find evidence that
the percentage poverty is correlated with ECE policy adoption as are the number of
working women in the state and the liberal ideology of state citizens. Interestingly, the
policy domain strength variable proved highly significant and increased the likelihood of
adoption by the highest log-odds. I present an analysis of the “reinvention” of ECE
policy by states in order to capture adoptions that happened subsequent to initial adoption
finding that early adopters of ECE were more likely to “reinvent” policy most likely to
adapt to changed policy expectations. Furthermore, reinvention policies were most likely

to involve teacher policy and funding. Finally, I looked at the usage of federal policies in

non-adopting states and find that most are located in the mountain plains and that they
have significantly higher rates of usage of Head Start and Special Education than their
peers that did adopt policy. While these findings show promise for understanding under
what conditions states adopt ECE policy, there are many factors that are not accounted
for in my models and which I hope to accommodate in a future expansion of this project.
Conclusions

The aim of this dissertation was to delve into the issues surrounding the early
childhood education. What is remarkable about this policy area is that despite an
enviable amount of support and momentum, there are major gaps in quality,
inefficiencies in policy overlaps and confounding definitions of the goals of the policy.
What is evident flom the findings presented herein is that the evolution of the policy has
depending greatly on the different problem definitions used by policymakers at different
times. Furthermore, flaming of this policy has wavered between flames used for welfare
and education policy. Two flames, however, seem to set ECE apart, the evidence flame
and the investment flame. Finally, there is evidence that at the state level partisan
politics has little to do with enacting policy, and that a state’s poverty level and level of
working women correlates with enactment. Reinvention in this policy area occurs
primarily in early-adopting states and is most likely to involve teachers and/or funding.
Non-adopting states utilize federal preschool programs in greater number than their
adopting peers.

ECE policy has certainly grown and changed over time. This body of evidence
raises a set of concerns in this policy area. First of all, quality concerns have been

present flom the get go and there still seems to be a considerable trade-off made between

10

funding and quality. Despite evidence flom the solid body of research that quality is
important to achieve documented outcomes; the same evidence that is professed in the
flaming of ECE is often ignored in the funding and implementation. An additional
concern is that while the original aims of Head Start were to address the massive
problems of poverty and racial inequality, these problems have not gone away and ECE
policy has not proved to be a simple panacea policy to solve them. The massive federal
programs of Head Start and CCDF continue to dominate public investment in ECE, and
while recent efforts to improve quality in Head Start show promise; CCDF continues to
operate largely as a custodial program for the very kids that research shows benefit most
flom high quality developmental preschool programming. An additional concern is with
teacher qualifications and training. Because the funding and quality trade-off has become
common, paying highly qualified teachers becomes tricky and as such teacher quality
remains a major concern in this area. At a time when public schools have focused
dramatically on “highly qualified teachers”, it is tragic that the same attention and
concern is not lavished on the early childhood profession.

The good news appears to be that states, the laboratories for policy innovation,
have been more successful at provisions of quality and at making program provision
universal. In addition, while public elementary schools certainly have a decent share of
the ECE programming, the majority of preschools are still privately operated. This could
mean that unlike kindergarten’s absorption into the system its makers were trying to
avoid, preschool will remain by and large a separate institution. Because states are also
the locus for implementing welfare and other social services programs, the possibilities

for coordination and comprehension of services is greater here.

11

More general concerns arise in the area of family policy and child development.
The notion of the family as central to ending poverty and to raising up blacks as a group
were central to 19603 notions of appropriate policy solutions. The ability of such an
amalgam of policies serving families in an efficient and sustaining way is unclear.
Piecemeal approaches to family policy are likely inadequate, and while ECE is not the
sole policy operating in that realm, it is certainly a major one. This gets to another
fundamental strain in this policy area: between home and family. With the prevalence of
non-parental care for young children, it is evident that it is no longer an Ozzie and Harriet
society. Yet some children do stay home a considerable amount of their young lives and
all children are at least home some of their young lives. As such, the developmental
concerns of policy must encompass the home and family in order for success to be an
option. The evidence that is cited so often that has taken on mythical proportions (Fuller,
2007), includes provisions for both parental involvement and home visitation. That
program element is perhaps the least understood and the most likely to be eliminated in
program implementation. The modest results seen in large-scale implementation of
public ECE policy should be carefully reconsidered. The history and development of
ECE policy, if nothing else, should caution policymakers to consider all elements of

quality when developing and reinventing policy.

12

Chapter One: Current and Historical Status of ECE

This section aims to present the status of young kids, of Early Childhood and of
government involvement in ECE currently and over time. This is primarily a summary
chapter based on national and state data that demonstrate the current standing and the
growth over time in this policy area. It also presents evidence of general consensus in
this area among political, ideological and public groups. Finally, I present less prominent
countertrends and dissent that are important to acknowledge in forming a complete
picture of this policy domain.

Current Status

There are currently more than 24 million children under age six in the United
States. This represents around 6.5 percent of the total US. population (U .8. Census
Bureau, 2008). These 24 million children come flom diverse backgrounds in terms of
race, ethnicity, income and family status. Around 55 percent are white, 14 percent are
black, 23 percent are Hispanic, 4 percent are Asian, and 1 percent are Native American
(U .S. Census Bureau, 2007). Children come flom families of varying financial means,
with 20 percent are at or below the federal poverty limit (FPL), currently $35,200 for a
family of three (NCCP, 2008), 23 percent are low-income (families earn between 100 and
200 percent of the FPL) and the remaining 57 percent flom families above low income.
This means that more than 43 percent of young children come flom families with low
incomes or families in poverty. Statistics for these young families tend to compound
each other as other risk factors associated with childhood are more prevalent in children
from low-income and poor families. For example, these children are more likely to have

parents with less than a high school education; more likely to live with a single parent

l3

and are more likely to move flequently due to displacement, eviction and guardianship
changes (NCCP, 2008).

A maj ority- 60 percent- of young children are in some type of non-parental care
on a weekly basis (N CES, 2005). The National Household Education Survey (NHES)
through the National Center for Educational Statistics (N CES) contains the most recent
set of national aggregate data on child care and preschool participation. Table 1.1
summarizes to types of care below. Non parental care is more likely in older children as
is center-based care (NCES, 2005).

Table 1.1: Care Arrangements for Children under 6

 

 

 

 

 

 

 

 

Age Non-parental Relative Non-Relative Center-based
care care care care

All under 6 60% 22% 16% 33%

Less than 1 40% 21% 14% 8%

l 53% 22% 20% 16%

2 59 23 18 25

3 65 22 14 43

4 79 21 13 65

5 82 20 13 73

 

 

 

 

 

 

 

Source: NCES 2005, NHES survey

Participation varies with other factors such as race, mother’s education, mother’s
working status and income. Black children are most likely to be in non parental care
(75%) than white children (60%) or Hispanic children (48%) (NCES, 2005).
Participation varies by income as well with fairly steady rates between poor (55%) and
low-income families (57%), but rising more dramatically in upper income families (72%
in highest income bracket). A mother’s level of education is also determinative of
participation with a sharp distinction between those with some college (63%), a BA
(65%) and post-BA (74%) and those with high school (56%) or less education

(43%).(NCES, 2005). Not surprisingly, mothers that work full time are most likely to

14

have children in non-parental care (85%) than non-working mothers (31%). The only
regional variation in non-parental care participation is in the West which has significantly
lower participation rates (54%) than the Northeast (63%), South (62%) or Midwest
(62%).

While much attention in ECE policy focuses on public programs, the fact remains
that around 60 percent of the funding for ECE is still flom families and private sources.
The remaining amount is flom federal (28%) and state sources (10 %) (Barnett and
Masse, 2002). Total spending in this area can be estimated at around $58 billion - the
sum of $16.4 billion federal, $7 billion state and $35 billion private spending.l Spending
per child varies widely with each program and for each family. Head Start spending is
around $6,700 per child (Head Start Bureau, 2007), and the average state investment (for
states spending on state preschools) is $3,600 per child (NIEER, 2007). Likewise family
spending on childcare varies. The SIPP reports that of the 4.9 million women who are
employed and have children under the age five, childcare expenditures are on average
$4888/year per child, which accounts for an average of 8.57 percent of their monthly
household income (SIPP, 1999). This cost proves to be regressive with those mothers
making the least (less than $1,500/month) paying the highest percentage of their income
(31.6%) towards childcare expenses (SIPP, 1999). As income is highly correlated with
education level, this is also true for the least educated mothers who pay 14 percent of

their income on childcare compared to the highest educated at 7.75 percent (SIPP, 1999).

 

I This is a rough estimate based on federal spending figures totaling $16.4 billion in 2006 and using Barnett
and Masse’s ratio of 60% private, 28% federal and 12% state. NIEER estimates $3.72 Billion were spent
on state preschools in 2007, and the remaining state money is for other ECE programs such as Head Start
supplements, CCDF matches, etc.

15

There are different types of care available to young children across the country.
Participation in each type also varies with age, income and race. Generally speaking, non
parental care is made up of three or four categories: center—based care (which includes
day care centers, preschools, and Head Start programs); non-relative home care (which
includes home-run day cares and group day cares); and relative home care (when a
relative watches a youngster either the child’s or the relative’s home). Over 33 percent of
young children are in center-based care (with the number rising to 73 percent for five
year olds). Around 16 percent of young children are in nonrelative home day care (this is
more common for one and two-year-olds — 20 and 18 percent, respectively). Finally, 22
percent of young children receive relative care with little variation based on age, but
much variation based on ethnicity (most likely for black children) (Mulligan, et al, 2005;
NCES, 2005).

ECE programs are funded and operate at all levels of government and in private
for-profit and non-profit settings. The largest single program is Head Start which now
serves more than 900,000 children each year with a budget of over $6.8 billion —roughly
$6,900 per child (Head Start Bureau, 2006). HS serves children ages three through five
not yet enrolled in kindergarten and provides support for their families (Head Start
Bureau, 2006). Head Start agencies, sometimes called preschools, provide a range of
services including education, early childhood development, medical, dental, mental
health, nutrition and parent involvement. It serves children flom families that are at or
below the federal poverty level (FPL), via its mix of educational programming for the

children and these support services for the children and families.

16

While Head Start serves as the dominant federal program for preschool
programming, federal child care programming has grown to now serve more children
than Head Start though with slightly lower federal dollars. The Child Care Development
Block Grant was enacted as part of the Omnibus Budget Reconciliation Act of 1990 (PL
101-508) establishing the primary source of dedicated federal fimding for child care
(Butler and Gish, 2003). The main purpose of this program is to send money via block
grants to states to be used to subsidize child care expenses for families that are considered
low-income or on welfare for their children under age thirteen (ibid). The CCDBG was
significantly amended and overhauled in this vein as a part of the Welfare Reform laws
enacted in 1996, most notably in the Personal Responsibility and Work Opportunity
Reconciliation Act (PRWORA) which established the Temporary Assistance for Needy
Families (TANF) to replace the existing welfare entitlement programs (PL. 104-193).
The CCDBG is now a combination of discretionary and entitlement fimds called the
Child Care and Development Fund (CCDF). What is most significant for ECE in the
revised CCDF is that for the first time, child care provisions were tied to welfare
recipients’ (and low-income, non-TANF recipients’) efforts at self-sufficiency. The most
striking feature of the TANF program was to allow states to establish time limits for
welfare recipients so that after a given number of years, persons are no longer eligible for
TANF funds, though no such time restriction exists in six states and no such time limits
are in the CCDBG law (Butler and Gish, 2003). Currently, there are over 1.8 million
children ages zero through thirteen that benefit flom the CCDF with approximately 60
percent of these or 1.1 million under age six (Child Care Bureau, 2006). Almost $5

billion was appropriated for the CCDBG in 2006.

17

In addition to Head Start and the CCDF — the two largest programs in terms of
funding and children served — there are range of other federal programs for childcare,
preschools and early childhood programs. These include a series of fimding streams
administered by two federal agencies: the Department of Education and Health and
Human Services (HHS). Programs fimded through Education legislation include Title I
Preschools, Early Reading First, Even Start, Special Education Preschool Grants and
Early Childhood Educator Professional Development Grants. In FY2005, funding for
these Department of Education Programs totaled almost $1.3 billion. Programs fimded
through HHS appropriations include Head Start, Early Head Start, the Child Care
Development Fund (CCDF), TANF funds and transfers flom SSBG. In FY2005,
appropriations for these programs totaled $15.1 billion (Schweinhart and Fulcher-
Dawson, 2006). Besharov and Higney estimate that since 1997 (first year of welfare
reform provisions) spending on these programs has doubled (Besharov and Higney,
2006).2

Currently, thirty-eight states fund and run forty-eight preschool programs
(Barnett, et al., 2005). Twenty states appropriate supplemental Head Start fimds, either to
serve more children or to support a range of things that the federal money does not
sufficiently cover, such as administration and coordination of training. For four states this
is the only state investment in early childhood. These state preschool programs range
flom full-day preschool for all four year olds to smaller, targeted half day programs for

poor children not eligible for Head Start (NIEER, 2007). .

 

2 Besharov and Higney use a slightly different set of federal programs to calculate this including the Child
and Adult Care Food Program (CACFP), but the funding increase is roughly equivalent.

l8

The National Institute for Early Education Research (NIEER) now publishes an
annual preschool yearbook with evaluations of preschool commitments of all fifty states
and has recently released its third such report (Barnett et al., 2006). As of 2004-2005,
NIEER reported that more than 800,000 children are served in the United States by state-
funded preschool programs. This represents 17 percent of all four year-olds and 3 percent
of all three year-olds nationwide. And it means that state preschools now serve almost the
same number of children as Head Start.

It is possible to estimate the number of children served by public programs, by
combining the numbers of each program above and comparing this to the national
number of young children. However, these are rough estimates as there is often overlap
in the children served by these programs (some children attend both Head Start and state
preschool programs). So a rough estimate shown in Table 1.2 reveals that around 17
percent of all children under age six in the United States go to public early childhood
programming. This number rises when you consider only children flom low-income and
poor families, where almost 40 percent of children receive government ECE

programming.

Table 1.2: Estimates of children served by major public programs.

 

 

 

 

 

 

 

Number of Percentage of all US Percentage of all low
children served children under 6 income and poor
children under 6
Early Head Start and 982,000 4 9
Head Start
State Preschool 935,000 4 9
Special Education 1 ,005,000 4 10
CCDF 1,114,500 5 11
4,067,000 17 39

 

 

 

 

Sources: Head Start Bureau, 2008; NIEER, 2007; US Department of Education, 2006,
HS, 2007; NCCP, 2007.

19

 

Historical Trends

In the past forty years, there have been simultaneous trends in policy and society.
As women increasingly worked outside the home, their child care needs increased and as
public firnding for child care and preschool increased more attention was paid to the
importance of learning in early years. In addition, as the promise of preschool learning
became more well-known, ECE as an issue area gained more champions and advocates
across the country. These growth trends and evidence of support for ECE are discussed
here in order to lay out the scope of the issue, the change over time and the appearance of
momentum and consensus. Subsequently, I lay out countertrends and evidence of major
shortfalls and disagreements in this policy area.

Beginning shortly after World War II, the number of women in the workforce
increased each year. In the 19603 and 19703 this trend continued, but what changed was
the number of women with young children (younger than six years old) began to increase
as well. By the mid 19803, 50 percent of all women with a child under age six worked
outside the home and by the late-19803, 50 percent with a child under age three worked
outside the home (BLS, 2006). Figure 1.1 shows the growth in female employment over
the last twenty years by the presence and age of the youngest child in the home (BLS,
2006). While the presence of all women in the workforce grew steadily, the rates of
women with dependent children grew the fastest with school-age children, children under
age six, and children under age three. Taken at face value this shows quite a clear

demand or need for some arrangement for the young children of a majority of women.

20

 

 

 

 

 

 

 

 

 

_ ___—_<

+ all women

 

 

+ no dependent
kids
kids under 18

 

 

 

‘ -—-~‘x-—- school age kid:

 

 

 

l' +kids under 6

Percentage of Civilian Workforce

 

 

 

 

 

 

' { +kids under 3
4° * — . " “mars
30 AI 1 I I I r 7 I I r I I T I I ‘7'”.7' _I'“‘T_fl I I I t I I I l j
(o ’\ Q N 'b ‘3 ’\ 0.) '\ ‘b ‘0 ’\ 0.) '\ “b

 

 

 

 

 

Figure 1.1: Working women by presence and age of youngest child: 1975-2004
Source: Women in the Labor Force: A Databook (BLS, 2005)

Not surprisingly, the growth in preschool enrollment tracks closely with the presence of
women in the workforce. Figure 1.2 shows the enrollment in preschool by age and also
shows the enrollment in kindergarten for purposes of comparison. It is worth noting that
at the time of Head Start’s enactment, less than 10 percent of all three and four year olds
were in preschool. But since then, enrollment has grown each year; in fact, this
enrollment has grown at a faster rate for threes and fours than women’s workforce
participation — due likely to the fact that many children enroll in preschool even if their
mother is not in the workforce. Unfortunately, such data for all ages in all arrangements

is not available for this time period, but the preschool enrollment data suggests that by

21

2003 almost as many four year olds were in preschool as five year olds were in

 

 

 

 

 

 

 

 

kindergarten.
90%
80% _,_ ,_ ,.
+3 year olds
70% l
60% l 4yearo|ds {
50% ~ ~
1 +53 in prek
40% - ~ 1
30% l +5‘s in
20% , _ . _ kindergarten l
10% «
0% . . . . , . .
1965 1970 1975 1980 1985 1990 1995 2000 2003

 

 

 

Figure 1.2: Historic Enrollment in Preschool and Kindergarten: 1965-2003
Source: NCES, 2003

While women’s workforce participation and preschool/child care enrollment has
grown, so too has the govemment’s role in ECE. The real origin of govemment-based
ECE policy is the passage of Head Start legislation in 1965. Initially passed as a
summer-only program for four, five and six year olds and their families in poverty, Head

Start has grown each year serving more children and their families (Vinovskis, 2002).

22

 

Head Start Spending: 1965- Present

8,000,000,000
7,000,000,000 -
6,000,000,000 -__.._._-._ W
5,000,000,000 4
4,000,000,000 l
3,000,000,000 .__.._. --___ . _-_ .-_.__
2,000,000,000
1,000,000,000 -

 

   
   
    
 
  
  

W + Appropriation (current '
dollars)

l—t—Appropriations (2002 l
#' "r dollars)

_._._._____________._ ___,_ ._—l

l

 

 

l

 

Spending

 

 

 

  

 

 

IIIJITIIITIIIIIII'T

\

$e§§3“§§§§§§§a§

 

 

 

Figure 1.3: Head Start Spending
Source: Head Start Bureau, 2006

 

Head Start Children Served: 1965 to Present

 

1,000,000
800,000 9
700,000 -
600,000 —
500,000 ~
400,000 ~
300,000 ~
200,000 ¥ ._ -._ -, W---" , ,

100,000 ._-_ _________, __--. W .._____. __ .

0 . . ,,,,,,,,,

\gbngPWANQK assessors» weer

.190

 

 

 

 

 

 

 

Figure 1.4: Head Start Children Served
Source: Head Start Bureau, 2006

Head Start funding (overall and per child) has increased roughly with inflation and while
the number of children served has plateaued around 900, 000 over the last three or four

years, other programs have been implemented and expanded during this time, such as

23

Early Head Start and Even Start which were added as complimentary programs to Head
Start serving children before they enter Head Start and families via adult and family

literacy programming (Good Start, Grow Smart, 2006). Figure 1.5 shows that per child
spending has followed 5 lightly different trend plateauing from 1974 to 1990, increasing

until 2000 and then plateauing again in the last few years.

 

 

Adjusted spending per child: 1965 to Present

7000 _ - W WW, f
l
5000 4 ~ . _. - _7 . z

4000 w
3000 —
2000 +
1000

0» ,
fey, 4~ ,4v ,0 ,49 ,4? ,4?» 4,9,0 ,40, 0%

 

 

(2002 dolla's)

 

 

 

 

 

 

0M0

 

Figure 1.5: Adjusted Head Start Spending per Child
Source: Head Start Bureau, 2006

Like Head Start, federal child care funding has experienced growth. Currently,
there are over 1.8 million children ages zero to thirteen that benefit from the CCDF with
approximately two-thirds of these or 1.1 million under age six (Child Care Bureau, 2006).
Almost $5 billion was appropriated for the CCBDG in 2006 which is more than four-
times the original amount for CCDBG in constant dollars3. In fact, as a part of welfare
reform through PRWORA, child care funding saw its biggest jump in federal to state

block funding because PRWORA repealed three older childcare laws via the old AFDC

 

3 The other three programs were AFDC Child care funded at $470 million, At Risk child care funded at
$270 million and Transitional child care at $113 million.

24

welfare rules and combined all funding into the CCDF. In 1990, funding for these three
programs and AFDC was around $1.5 billion, so that current CCDF funding still reflects
a doubling of federal expenditures on childcare in the last fifteen years (Butler and Gish,
2003)

While Head Start programs were created to serve children from poor and low-
income families and other families increasingly sought out preschools for their children
in private arrangements, state governments were not initially involved in preschool
policy. With the exception of California and New York where policies were developed
and enacted simultaneously to Head Start policy (in 1965 and 1966 respectively), other
states played at most a supporting or supplemental role in public preschool. However,
beginning with Florida in 1978, states began to develop and enact preschool policies that
did more by targeting specific populations or requiring certain quality standards (N IEER,
2005). Over time more and more states added state funding to the mix of federal and
private offerings for preschool. Figure 1.6 demonstrates the growth in the number of

states with state preschools over the last forty years.

25

 

 

 

 

 

Number of States wuth State Preschool

40

35«

301

25«

20a

15.

10~

5,

0 FITTIITfi 1111 171 :1,
«DOOPngl-OCONmODOFNCOVIOCONQGOu-
txrxooooooao QQQQQOOOOQQOQOQOO
OQQQQQQODGOJQODODOJQOGGOODOOOO
FFFFf-FFPFPT'FFFFFFFFFFFNN

 

Figure 1.6: Number of States with State Preschool Policy 1978 - 2001
Source: NIEER, 2006; ECS, 2005

In examining these trends it is quite clear that there are positive historical trends
overtime both the need for (women working), the increased usage of (enrollment) and the
governmental involvement in ECE. There does, however, seem to be a plateauing in
more recent years a subject I return to later in this chapter.

Areas of Consen_su_s

Perhaps the most convincing element of the current ECE policy scene today is the
evidence of research itself. This comes from two separate fields of study — one hard
science (neuroscience), one social science (educational evaluation of programs) - that
now converge along several concrete findings. In addition, several major federal reports
and one international one have been issued that summarize the findings of this research,
but also highlight the growing knowledge of best practices in ECE.

With improvements in technology, the field of brain research, or neuroscience,

has been able to literally Show (via such things as positron emission tomography and

26

 

 

MRIs) the effects of different environments on children’s brains. A separate line of
research looks at children with localized brain damage (National Research Council and
Institute of Medicine, 2000) and identifies the conditions that are dangerous to
developing brains. The major findings are that young children’s brains are harmed by
such things as abuse, neglect, and harsh environments.

In a parallel set of findings, there is a set of three studies of ECE programs - The
High/Scope Perry Preschool study‘, the Carolina Abecedarian Project studys, and the
Chicago Child-Parent Centers study6- that are widely cited for their longevity and design
quality. They offer consistent and valid evidence of both the short-term and the long-
tenn effects of quality preschool programs. While the studies varied slightly in design,
scope and program elements, their findings converge along a common dimension: ECE
programs result in immediate and long term benefits to the child especially in terms of
academics7. Participants saw significant gains over non participants on such things as
graduation rates, school achievement test scores and decreases in such things as special
education placement, retention and dropping out (Schweinhart and Fulcher Dawson,
2006). Beyond educational benefits, these model ECE preschool programs have

significant impact on future economic status (income) and criminal behavior (decreased

 

4
5

See for example, Schweinhart, Montie, et al.
See for example, Campbell and Ramey or Clarke and Campbell
6 See for example, Reynolds, et a1.

7 From Schweinhart and Fulcher-Dawson, 2006): “Programmatically, the Abecedarian program was a full-
day child care program serving children from shortly afier birth to kindergarten entry, while the
High/Scope and Chicago programs provided part-day preschool education to children 3 and 4 years old and
related services to them and their families. The Abecedarian and Chicago programs provided a range of
family and health services, but only the High/ Scope program provided weekly home visits to parents and
their participating children. The High/Scope project provided no follow-up program in elementary school,
while both the Abecedarian and Chicago programs did provide school-age programs, although the
Abecedarian study did not find these school-age programs to contribute to children's success as did the
preschool child care program in the same study.”

27

Because these programs all served poor and low income families and their
children, a great deal of attention has been paid to the economic impact of these
programs. The economic impact is calculated based on the savings to taxpayers on things
such as welfare, criminal incarceration, grade retention, and taxes paid on higher
earnings. Based on the program costs, and the future cost savings on such societal
spending, the cost to benefit ratio for each ECE program has been calculated with the
Perry Program at $17 saved for each $1 spent, Chicago at $7 and Abecedarian at around
$4 (Schweinhart and Fulcher Dawson, 2006). These cost benefit calculations have
compelled several top economists to argue that the benefit of ECE programs is stronger
than the evidence for most other public investments (Heckman, 2006; Rohiick &
Grunwald, 2003).

In addition to these three well-cited and well-regarded longitudinal studies of
quality preschools, several national and international reports have been issued in recent
years offering compendia of the various research and evidence in all fields pertaining to
early childhood. The National Research Council and Institute of Medicine prepared an
excellent report on the research from neuroscience, From Neurons to Neighborhoods: the
Science of Early Childhood Development.i The National Research Council Committee
on Early Childhood Pedagogy published the book, Eager to Learn: Educating our
Preschoolers, which “represents the first attempt at a comprehensive, cross-disciplinary
synthesis of the theory, research, and evaluation literature relevant to early childhood
education” (National Research Council, 2000). The authors state that, “as the twenty-

first century begins, there can be little doubt that something approaching voluntary

28

universal early childhood education, a feature of other wealth industrialized nations, is
also on the horizon here” (National Research Council, 2000).

In addition, the IEA Preprimary Project examined various types of early
childhood settings and their relationship to child outcomes, not only in the US, but also
in other countries around the world. It was sponsored by the International Association for
the Evaluation of Educational Achievement (IEA) and coordinated by the High/Scope
Educational Research Foundation (Montie, Xiang, & Schweinhart, 2006; Olmsted, &
Montie, 2001; Weikart, Olmsted, & Montie, 2003). The study examined 1,300 to 1,897
children from age four and one—half to seven in ten countries, with an overall retention
rate of 86 percent of the original samples in the participating countries to see how process
and structural characteristics of community prepn'mary settings affect children’s language
and cognitive development. Across these different intemational settings, researchers
found that children’s language performance at age seven improved when children freely
chose their activities, when the teacher had more full-time schooling, when children did
less whole-group activities and when there were a greater number and variety of
equipment and materials.

Beyond the compelling evidence coming out of various research disciplines, there
is a great deal of consensus politically for ECE policy. Head Start has largely developed
as a bipartisan or possibly nonpartisan issue which parallels the increases in government
involvement in child care and preschool under a variety of partisan and unified or divided
government make-ups. Every president since LBJ (except Nixon) has mentioned
activity on policy in either Head Start, preschool or childcare in at least on of their State

of the Union addresses (Woolley and Peters, 2008). The major party platforms

29

(Democratic and Republican) increasingly attended to the policy issues and necessities
for ECE (Woolley and Peters, 2008). This is most remarkable in the Republican party
which initially (early 19708 particularly with Nixon) was against government
involvement in or support of child care claiming it was a private, family matter, but in the
late 19705 and early 19805, the Republican party switched tacks and began to endorse
childcare and preschool spending with the condition of increased privatization and
allowances for faith-based organization involvement (Cohen and Sardell, 2002).

As mentioned before, states are increasingly seen as the locus for policy
innovation and development for ECE. In 2007 alone, nineteen of the governor’s giving
speeches spoke of ECE in their State of the State Addresses. Since 2000, more than
seventy different ECE initiatives and laws have been acted on in the states (ECS, 2007).
Both the National Governor’s Association has been a strong voice for both aligned
systems of ECE and “have taken the lead in recognizing each state's fundamental
responsibility for a seamless progression in education for citizens from their earliest years
through college and into lifelong learning” (NGA, 2007). The National Conference of
State Legislatures supports early learning and a more comprehensive approach to ECE
policy via improved federal-state interaction (NCSL, 2007). What is perhaps most
remarkable is that all of these state efforts have been either bipartisan efforts or have been
supported in different places by different parties reflecting the reality at the federal level
that ECE is largely a non-partisan issue.

In parallel with these government and political supports for ECE, there has been
growth and coordination among the ECE advocacy and interest groups across the nation.

Nationally, there are a set of strong, prominent advocacy groups that increasingly support

30

various ECE policy efforts at the state and national levels. Some usual suspects have
emerged as prominent advocates whose voices and policy positions are widely heard
including the Head Start Association (and its affiliate state groups), the Children’s
Defense Fund(CDF), Birth to Three, the National Association for the Education of Young
Children (NAEYC). In particular, the CDF is widely cited as a major influence in the
growth of advocacy organizations and in the coalition of them to advocate policy in
Washington (Cohen and Sardell, 2002). Between the 19708 and the late 19808 the major
coalition for child care grew from twenty organizations to more than one hundred largely
due to the CDF’s leadership (ibid). In addition, the influential NGA and US. Catholic
Conference came to be pivotal supporters of ECE through their child care advocacy
efforts in the 19808 and 19908 (ibid). For the most part these advocacy groups come
together to support major initiatives in ECE though they may have particular elements of
the policies that are more pertinent to their own groups. What results is a great deal of
consensus and coordination in ECE lobbying and advocacy efforts. Recently, the UPK
movement led by the Pew Charitable Trust has gained prominence in its support of state
universal preschool efforts.

In recent years, a group of advocates have emerged to champion various ECE
initiatives that are not “insiders” in the ECE world. These are both individuals and
outsider groups. For example, Rob Grunewald of the Federal Reserve Bank of
Minneapolis and James Heckrnan, Nobel Laureate of Economics have both in recent
years become vocal advocates for investing in Early Childhood Education largely based
on evaluations of the striking economic return for such investment in the model preschool

programs mentioned above. In addition, more well-known national figures such as the

31

actor Rob Reiner who championed California’s recent attempt at Universal preschool and
First Lady Laura Bush, have been visible supporters of ECE. Outsider groups parallel
this as they come largely from business sector, such as the Business Roundtable and the
Economic Policy Institute who have both promoted increased attention and funding for
ECE.

Countertrends?

While there is a body of evidence supporting the momentum and consensus View
of ECE, there is also a set of information that is less encouraging, but equally
informative. As such, I now turn to what I call “countertrends” that arguably are more of
a “glass-half-empty” viewpoint, but this is crucial to fully realizing the accurate
landscape of this policy area, rather than just the pretty picture often presented
incompletely. Most notably, the areas of poor children served, spending instability,
overall quality ratings, CCDF quality concerns show discouraging and downward
numbers. Most prominent are two other issues: teacher quality issues and evidence
issues.

First of all, given that many ECE programs aim to help poor children, in
particular, because of the known negative consequences of a childhood of poverty, it is
discouraging to know that only around forty percent of all poor and low-income children
participate in public ECE programs. This means that a fill] 60 percent of young children
are not benefiting from knowledge of early learning or the public will to help them.
Furthermore, private or family spending on ECE is highly regressive with families in the

lower income brackets paying the highest percentage of their income toward childcare — a

32

Catch Twenty-Two situation in which families earn too much to qualify for federal or
state programs, but make little enough that ECE expenses are particularly burdensome.

A second and related point then is to the issue of funding which drives the number
of children being served and the quality of programming. While funding for larger
federal programs remains high, it has plateaued in recent years as is most evident in the
Head Start and CCDF appropriations failing to keep up with inflation in the last five
years. More concerting are state preschool funding streams which seem particularly
susceptible to public manipulations and budget cuts. Afier controlling for inflation,
eleven states had lower spending on preschool programs in 2004-05 than in 2001-02;
eighteen states had the same amount of funding. In constant dollars, overall spending per
child decreased in the US. by 7.3 percent from 2001—02 to 2004-05.

Overall quality of ECE programs remains an issue across the board. Recent
efforts have surely been aimed at Head Start to increase the number of teachers with
BA’s and to assess achievement in participating children, but no radical changes have
been made such that quality concerns remain. States have been slowly improving the
quality of existing programs, though the progress has been slow and unsteady, with
inconsistent funding and changes in program definitions. New programs that states
implement tend to have higher quality ratings, though not always. Currently, there are
twenty-two states whose programs have a quality rating of six or higher (on the NIEER

ten—point scales), sixteen states have scores between one and five, and twelve states have

 

8 NIEER ranks states in three ways: access to preschool, quality and resources. The quality ranking
depends on meeting a checklist of ten factors that includes: early learning standards (comprehensive),
teacher degree (BA), teacher specialized training (specializing in prekindergarten), assistant teacher degree
(CDA or equivalent), teacher in-service training (at least fifieen hours/year), maximum class size (twenty
or fewer), staff —child ratio (1:10 or better), required screening/referral and support services (vision,

33

no scores due to lack of program. It is worth noting that the NIEER ratings and rankings
are not scaled or weighted, for example, serving a meal at a preschool is counted and
valued the same as having a well-qualified and trained teacher, even though evidence
supports the value of teacher quality quite highly. Only Arkansas meets all ten
benchmarks, and only five other states meet nine of the ten. It is of concern that of the
thirty-eight states with programs, twenty-one of them meet five or fewer of the quality
benchmarks.

The CCDF program has certainly grown to serve more children with more money
since its original incarnation eighteen years ago; however, quality is still largely an
afterthought as the main outcome goal of CCDF — under the umbrella of all welfare
reform — is to get families off welfare. As such, jobs for parents is the number one goal,
and care for children while their parents work is a subsidiary goal with less attention to
quality than custodial care. This is nowhere more evident than in the fact that almost 30
percent of all CCDF care providers are unregulated and even regulated providers have
minimum regulations aimed at health and safety concerns, rather than developmental,
social or academic quality concerns (Child Care Bureau, 2006). This means that more
than 200,000 fiiends, families and neighbors (FFN) are given public fimding to care for
children regardless of any true credential or training to do so (Child Care Bureau, 2006).
Major countertrends: teacher quality and short-term evidence

Perhaps the most discouraging trends and current information in this policy area
center around the teachers who are most critical to making the programs work. The

National Association for the Education of Young Children (NAEYC) released a report

 

hearing, health; and at least one support service), meals (at least one a day), and required monitoring (site
visits).

34

containing the “Critical Facts about the Early Childhood Workforce” based largely on
data from multiple sources including the US Department of Education, the National Child
Care Information Center (NCCIC) and the Center for the Child Care Workforce. As
reported in this brief, there are more than 2.3 million paid teachers and caregivers for
children ages zero to five. Of these 24 percent are in center-based settings, 28 percent are
family childcare providers, 35 percent are paid relatives and 13 percent are paid non-
relatives (e. g. nannies) (NAEYC, 2002).

The training requirements for each type of early childhood worker vary widely by
type of worker, setting and by state. For child care centers, thirty states have NO pre-
service training requirements for new workers, while eleven states (numbers include the
District of Columbia and New York City separately) require it of lead teachers, twenty-
two of teachers and forty require it of center directors. Most states require some form of
annual professional development (PD) for current workers (forty—five for teachers, forty-
three for directors). For family child care (home providers) there are fewer states with
requirements — preservice training is required only in eleven states and only thirty-four
require ongoing PD. (N AEYC, 2002). As of 1995, 80 percent of teachers working in
centers had some college education, with one—third having bachelor’s degrees. For
family home day cares, approximately 55 percent have some college-level education
while only 17 percent have Bachelor’s or more. The childcare workers average hourly
wage is $7.43 while preschool teacher’s average is $8.56. The average turnover. rate for
all early childhood workers nationally is 30 percent annually (NAEYC, 2002).

In addition, to concerns of teacher quality are concerns about the evidence of ECE

outcomes. While the noted long-term studies such as High/Scope Perry Preschools,

35

Chicago Child Parent Centers and Abecedarian, demonstrate clearly the major impact of
high quality ECE programming for young children, the evidence from short-term studies
of the larger-scale public programs that aimed to replicate these model programs is not so
clear. Generally speaking publicly funded ECE programs have been shown to improve
literacy and social skills of children and have positive effects on parental behavior
(Schweinhart and Fulcher-Dawson, 2006). These scaled-up programs like Head Start and
State preschool generally have more modest and short-term effects on the children than
the model programs showg. For example, in the F ACES10 study children in Head Start
gained on average four points on the Peabody Picture Vocabulary Test while children in
Perry preschools gained on average eight points on the same test which is a twofold
difference (Schweinhart and Fulcher-Dawson, 2006).
Conclusions

The current status of ECE demonstrates a great deal of activity in this issue area.
And while most historical trends point to overall growth over time in terms of supply and
demand of ECE, this masks a great deal of inconsistency and instability in this issue area.
While programs continue to receive funding and new programs are enacted, quality
provisions are still dropped in lieu of funding and access concerns. In order to save
money and/or serve more children, teacher pay and training, classroom size and length of
program are cut despite evidence that these are crucial to the outcomes in ECE. In

addition, solid long-term evidence used to garner support for policy is hampered by short-

 

9 There are several short-term findings from a set of studies that includes: the Head Start Impact Study,
FACES, an evaluation of Early Head Start, the Head Start Comprehensive Child Development Program
Evaluation, evaluations of Even Start and Barnett, et al.’s examination of the effects of five state
reschools.
0 The Head Start Family and Child Experiences Survey (FACES) was done with two cohorts entering HS
in 1997 and 2000 totaling 6,000 children.

36

term evidence of actual to-scale public implementations supposedly based on these model
programs. Recent data suggest a plateauing of enrollments, percentages of working
women and funding suggesting, though not in any way conclusively, that the size and
scope of this issue has reached a peak or a stopping point. Percentages of children in
non-parental care do not seem likely to climb significantly higher causing demand to
remain steady. It is possible then that if access is truly close to maximized, quality can be

more fully addressed.

37

Chapter Two: {Merstaldjgg ECE Policy- a Theoretical Proposition

Given this picture of the past and present of ECE, I now turn to theory to establish
a framework for answering the questions at hand: why is ECE policy the way it is? Why
is it so varied? There are several ways to go about responding to these questions. Over
time, several theories of policymaking have been developed and empirically tested
producing a menu of options for those seeking to understand how, why and when policies
are made and to what effect. For purposes of understanding ECE policy, and more
specifically, why there is the current variation and history in this policy area, a few such
theories seem appropriate for examination. The attendant concerns of this paper are to
understand variation, particularly with regard to time and policy “space” or domain. As
such, theories examined here are chosen specifically because they attend to one or both of
these concerns.

Change and development over time is a particularly important aspect of
policymaking. Victor Hugo proposed that “Greater than the tread of mighty armies is an
idea whose time has come” which presumes both that ideas arise and that the idea’s
resonance in a community is dependent on time. ECE has a relatively “short” history in
the United States, a fact which naturally limits study of it to the last forty years (though
some might argue that this should be the last 150 years since the infant school movement
in Massachusetts took place briefly in the early-mid 18008). Yet even in this short time
period there have been both gradual and dramatic changes in this policy area.

Change and development over space or in policy terminology domain is equally
important in policymaking. This notion is perhaps more ambiguous than time in that it is

measured in terms of differing areas of government activity, different values people have

38

and in different goals for the modern state. Yet it is measurable and extremely important
to understanding how and why policies develop in variable ways. While time may make
certain changes seem inevitable at the incremental level, understanding the values,
government activity and goals in policy domains makes it possible to explain non-
incremental change and variation.

Three theories are presented here as separate, though related theories of
policymaking. As much as possible, they are presented in a way that allows for both a
comparison and a combination of the four in order to then thoughtfully apply them to
ECE. As such, this examination presents for each theory: the type of theory it is,
assumptions and need for it, elements and definitions, mechanics, evidence and
counterclaims. I have divided them into two types: two that attend to the policymaking
process and one that attends to policy development over time. In the Appendix, Table
2.1 contains a tabular comparison of the three. Finally, I propose how each is suited to
helping with the enterprise here: to heighten understanding of the variation and policies
ofECE.

The Policmaking Process: Frames, Streams and Windows

A critical examination of any policy begins actually prior to the policy’s
enactment —- during the policymaking process. John Kingdon’s work on policy streams
and windows is one of the most prominent in modern day policy analysis and theory.
Kingdon’s basic premise is that policies depend on a confluence of ideas, people and
events coming together at a critical time (window) leading to placement on the agenda.
A small part of Kingdon’s theory has to do with the flaming of a problem (problem

definition) and the resultant solution frames. While framing theory is one element of

39

Kingdon and streams theory, it is considered here more fully as it has been treated by
sociologists (and more recently some political scientists and educationalists). The reason
for this is that framing theory focuses on both what is considered (what’s in the policy
frame) and what’s not (what’s outside of the frame and therefore not considered valid for
a given policy discussion). In this way, framing is a pivotal part of policymaking. It is
not an easily captured phenomenon, however, and yet as will be seen in subsequent
chapters it is certainly measurable and empirically present in policymaking processes.
Multiple Streams, the Garbage Can Model and Kingdon

Multiple Streams (MS) theory is one that aims to explain policy formation by
looking at policy choices under ambiguity (Zahariadis, 1999). It is an attempt to answer
questions of how policymakers’ attention is rationed, how issues are framed and
how/where solutions and policy problems are searched for (Ibid). John Kingdon
contributed fundamentally to this theoretical field by expanding the “garbage can model”
of policymaking to the US federal government. He particularly illuminated two elements
of this process: agenda setting and alternative specification. His theory focuses on the
system-level of analysis (either the entire system is studied or a separate decision).
Kingdon asks how information affects choice between alternatives, that is, how inputs
transform into outputs (Kingdon, 1995).

MS begins with the ambiguity that policymakers face in their decision matrices.
Ambiguity is the reality that they are faced with many competing and non-reconcilable
ways of thinking about the same problem or situation (March and Simon, 1994). In
addition, policymakers must act within a specific time frame and are generally focused on

more than just the single issue at a time. The garbage can model developed by March,

40

Cohen and Olsen proposes that individuals cannot process all this information rationally
selecting a “correct” solution (March, et al., 1972). Rather there are actually limited
bands of choice (between alternatives) and the problem definitions are vague and shifting
(Zahariadis, 1999). Kingdon expands on the garbage-can model and focuses on agenda
setting. His theory makes it possible to explain non-incremental change and non-rational
decision-making. MS is important and unique because "it seeks to explain policy
precisely when the assumptions of clarity and self-interest are inappropriate descriptors"
(Zahariadis, 1999).

Kingdon argues that there are three elements of the policymaking process that
come together to produce policy: problems, policies and politics. At critical points or
windows in time, all three are put together, usually by a policy entrepreneur in such a
way that it increases the odds of attention and agenda placement. Problems are defined
either by statistics that demonstrate the magnitude of an issue; by dramatic events or
crises, or by feedback from an existing program. (Rochefort and Cobb, 1994). Policies
are made up a mixture of ideas that swirl around in a policy “primeval soup”, with only a
few ever actually chosen and enacted (Kingdon, 1995). Policies are selected based on
a) technical feasibility-whereby the more difficult they are to implant, the less likely they
are to be considered or b) value acceptability— whereby the further they are from
policymakers’ values, the less likely to be considered (Kingdon, 1995). Politics is made
up of three elements: the national mood, pressure group campaigns and administrative
and/or legislative turnover.

These three elements interact to lead to policy creation or reinvention. Two

concepts are critical to the production of policy in this model: coupling and policy

41

windows. Policy windows are critical moments in time when all three streams (problem
definition/s, policies and politics) are joined together in a process Kingdon refers to as
“coupling”. Coupling depends on the presence of policy entrepreneurs-be they
governors, lobbyists, celebrities- who must attach problems to their solutions and get
politicians to agree. As Kingdon notes, “solutions come to be coupled with problems,
proposals linked with political exigencies, and alternatives introduced when the agenda
changes” (Kingdon, 1995, p. 173). Windows - during which coupling must be done - are
short and either predictable (like an annual budget allocation) or unpredictable (like an
earthquake). It is precisely during this part of policymaking that non-incremental change

and seemingly non-rational decision-making occurs.

MS focus: Agenda setting + alternative specifications

Streams Agenda items Alternative (actions)
Problem HM
Policies A A1
Politic B V2
/ C 3
Framin

 

 

 

 

Policy Window
(Where entrepreneur does coupling)

Figure 2.1: MS Theory and Kindgon

While Kingdon made major advancements to help clarify the field of MS theory,
since his seminal work, additional tweaks have been made to the theory. First of all,
Kingdon’s unit of analysis was the national government and its multitude of issues, but
MS can also be used to look at a single issue (Zahariadis, 1999). In addition, Zahariadis

proposed that the three elements of the politics stream — national mood, pressure groups

42

and tumover — can accurately be combined into a single variable of ‘ideology of
governing parties’ (Zahariadis, 1999, p. 79). And while Kingdon’s laid a foundation for
improved work in this field, he focused on only the predecision process, while MS looks
at the whole policy process.

In addition to these modifications of Kingdon, there have been several challenges
to MS theory that continue to arise for researchers utilizing this theory for policy
analysis. One such challenge is the independence of the streams, for there is evidence
that the streams interact at times other than just during open windows. In addition, the
role of the “window” in coupling is imprecise: some windows are open in problem
streams, others in politics streams. Zahariadis argues that it depends on where the
window opens - if in the problem stream, policymakers want a solution to fit a problem;
if in the politics stream, they want to find problems to attach to their preferred solutions
(Zahariadis, 1999). Furthermore, some issues always have open windows, so the
projection that policymaking is dependent on a small finite coming-together is inaccurate
at times. In contrast, to these critiques, there is empirical evidence that MS does at least
predict that the ideology of political parties is important and that bureaucrats shape
solutions but do not change the agenda (Kingdon, 1995).

Issue Framing Theory

Issue framing theory (IF) - comes not from political theory, but from sociology,
particularly, the theories of social movements. IF stems largely from Goffinan who
argued that frames help people discern what is “real” within a given context (Goffrnan,
1974). This theory suggests that the crux of social movements is in how an issue is

framed; the frame focuses the public and the debate on certain things and thus renders

43

other facts and arguments moot (Schon and Rein, 1992). In other words, issue frames
provide people with boundaries of values within which they should consider and act on a
policy.ll Framing is then an element of Kingdon’s conception of agenda setting and of

alternative specifications.

Problem definition

\ Issue A:
Facts and Values

NOISE
(FACTS AND VALUES NOT CONSIDERED RELEVANT)

 

 

 

_ .1. a
--. ...-

 

 

 

 

 

 

 

 

Figure 2.2: Framing and Problem Definition

The need for flaming arises when there are policy disputes. There are two kinds
of policy disputes: policy disagreements (PDs) that can be settled by reasoned discourse
(appeals to fact and evidence) and policy controversies (PCs) those that are stubbomly
resistant to resolution via the exercise of reason (Schon and Rein, 1994). Facts play quite
different roles in PCs than in PDs. In PCs, policymakers pay selective attention to facts,
give different interpretations of facts and contestants discount opponents’ evidence.
Framing is what is needed to move a policy controversy forward in a meaningful way

when appeals to facts do not work. Frames focus attention on a few resonant features of

 

H Some concepts and theories have evolved and been establish in the flaming literature that are
defined here for sake of clarifying some of the terminology used in this research. An issue frame is a set of
ideas and values used to understand an issue (Gamson, Schbn and Rein; J acoby). Framing is the use of
flames to move policy/society forward on an issue (Snow and Benfield; Davies). Framing eflects are the
effect of flames on policy outcomes, implementation and public opinion (Iyengar; Nelson et a1). Frame
analysis is the retrospective study of flames in an issue area with positive conclusions (Goffrnan, Gamson).
Framing policy is the use of flames as a tool to get a policy enacted and implemented with normative
implications (Schon and Rein, Coburn).

44

an issue that is otherwise too complex to comprehend. “Through the process of naming
and flaming, the stories make the ‘normative leap’ flom data to recommendations, flom
fact to values, flom ‘is’ to ‘ought’” (Schon and Rein, 1994, p.23). In this way a familiar
set of ideas moves into a new situation, highlighting the problem and what must be done
about it.

So when there are policy controversies, flaming is needed to move an issue
beyond the gridlock of conflicting frames, but there must also be political opportunity to
do so (Davies, 1999). The opportunity for flaming to “work” and move an issue forward
then depends on changes in politics and society external to any flaming activities.
Political opportunities lead to a “tipping” point in a policy area where framing becomes
possible and quite important in getting an issue on the newly defined or open agenda
(Kingdon, 1995). The two major types of political opportunities are: dramatic events and
cultural shifts (Davies, 1999). It is in this way that flaming activity is quite dependent on
time and chance as it works primarily when there are major events and/or there are major
shifts in culture.

Frames have quite particular and specific attributes that distinguish them flom
such things as fabrication, political persuasion, “spin” and mere rhetoric. Frames are not
merely semantic, nor are they merely words or matters of the mind (how one thinks about
an issue), but also “affect the way an activity is organized especially for its main social
agents” (Gofflnan, 1974, p.247). This is crucial to the conceptual presentation of flames
—that they promote and guide both ideas and appropriate action.

Goffrnan set forth a set of five fundamental assumptions about flaming (Goffrnan,

1974). First, it involves systematic transformation of an issue that is already meaningfirl

45

in a different way. Second, participants are meant to know and recognize that an
alteration is involved in flaming. Third, cues establish the beginning and end of a
transformation. Fourth, flames are not restricted to certain types of events or classes of
perspectives. And fifth, sometimes flames do not alter an activity dramatically, but may
alter the fervor with which one acts or for what reason. (E. g. if you call checkers a
fiiendly game, the aim is fun, if you call it a competition, the aim is winning).

There are two ways of classifying flames: by the purpose of the flame and by the
values and norms appealed to. Snow and Benford provide a typology by purpose (Snow
and Benford, 1986): diagnostic, prognostic and motivational. Diagnostic frames both
define the problem at hand and assign blame for the problem. Such flames intentionally,
and unintentionally, focus on some aspect of the problem and not others. Prognostic
frames are solution oriented, in that they include a suggestion of how a problem may be
solved, and provide goals and tactics for achieving them. Beyond the diagnostic and
prognostic flames is the “motivational frame” which is a call to arms, a rationale for
engaging in action. Gamson calls this the agency component of flaming (Gamson, 1992).

Rein and Schoen highlight a typology of frames - based on the values and norms
they embody - delineating three types of action flames - policy flames, institutional
action flames and metacultural flames (Schon and Rein, 1994, p. 33). Policy frames are
those that institutional actors use to construct the problem of a policy situation.
Institutional action frames are generic flames institutional actors use to construct policy
flames. Metacultural frames are broad, culturally shared systems of belief (institutional
action flames are local expressions of these, e. g. disease v. cure; natural v. artificial;

wholeness v. flagrnentation). “Metacultural flames, organized around generative

46

metaphors, are at the root of the policy stories that shape both rhetorical and action
flames “(Schon and Rein, 1994, p. 34).

It is important to consider the effect (resonance) of flaming on targets
(constituents, mass publics, policymakers, practitioners). Resonance, or appeal of the
flame to targets, depends on the consistency of the flame, the empirical credibility of the
evidence, the credibility of the flame articulators and the salience of the issue and values
with the targets (Snow and Benford, 2000). This gets at the psychological elements of
flaming, studied in depth in the field of psychology, but covered only cursorily here. The
work by Nelson, et al will suffice to demonstrate the effect of flaming in individuals.
They found that “flames are distinct not only at the message level but at the
psychological level as well; that flames differ flom other message forms not just in their
overt structure and substance but also in the way they affect popular thinking about
public affairs” (Nelson, et a1, 1997, p. 222). This can be represented by a simple
equation A: Xviwi where A is the summary attitude, v is the value of attribute i.
(individual’s belief about the attitude object) and w is subjective weight of that belief.
Attitudes can be arguably changed either by changing individual’s beliefs about the
attitude object (persuasion) or by changing how the individual weights that information
(flaming). Thus flaming affects the personal importance one places on different
dimensions of an issue.

One last element of flaming that must be attended to is that of the “flarner”. The
literature suggests that flaming can occur at both the organizational and the individual
level (Snow and Benford, 2000). As such, flaming happens not only when political elites

attempt to garner support for an issue and gain votes, but also during the discourse

47

involved in political discussion (Gamson, 1992) and policy implementation (Coburn,
2001). While flaming may be done by elites and plurals, the activities and effects of each
are often different, though the literature does not currently address these differences.
Policv over Time: Diffusion and Innovation

While the first two theories focus on the predecision part of the policymaking
process and generally apply to a single policy or issue at a time, diffusion and innovation
theory addresses how it is that policies change and are enacted over time. This is the
notion that policy diffuses through different institutions over a certain period of time.
Diffusion and innovation theory focuses on the process through which govemments
adopt new programs (Berry and Berry, 1999). The theory suggests that there are variables
that must be considered when examining why a government enacts a given policy - in
particular, state policy enactment depends on variables internal to the state as well as
variables external to it (such as other states’ adoptions, federal policy, national economic
trends). This theory is important because it aims to explain what factors lead to policy
adoption.

This literature is rooted in long-standing, but separate theories for policy
innovation and for diffusion, but has begun to reflect a more thorough and unified
framework for empirical analysis. Based on the early works of Walker, Gray and Dye,
researchers segmented the field into two parts: the study of diffusion models and the
study of determinants models. However the theoretical underpinnings for diffusion of
innovation have seen major change and improvement in the past decade. Most notably
the work Berry and Berry have done using event history analysis (EHA) utilizes both the

regional diffusion and internal determinants models in a unified way.

48

Policy innovation is defined in the literature as any policy that is “new” to the
state adopting it (Walker, 1969; Berry and Berry, 1999). Internal determinants models
essentially argue that it is the political, economic and social characteristics of a state that
determine when and if it will adopt a policy innovation. Diflusion models, on the other
hand point to factors external to the states that affect such adoption. These models are
the national interaction, regional diffusion, leader-laggard and vertical influence models.
The difference between the models, according to Berry and Berry, is “the channels of
communication and influence assessed to exist” (Berry and Berry, 1999, p. 171). For
example, the regional diffusion models assume that states communicate with and are
most heavily influenced by the states in proximity to themselves while the leader-laggard
model assumes that there are certain states that other states look to for information and
leadership when it comes to policy innovation.

Berry and Berry moved the policy innovation and diffirsion field forward by using
event history analysis (EHA) to test their combined model of state innovation and
diffusion. Berry and Berry use a different dependent variable than had traditionally been
used in the field: a hazard rate, which is the probability that a state (i) will adopt a policy
during a time period (t). Using EHA, they define a “risk set” which consists in any given
year of the states “at ris ” for adopting the policy (essentially N minus the number of
states who have already adopted it). The advantages of this model are that it allows
researchers to test regional (diffusion) and internal (determinants) factors, it can account
for factors that vary flom year to year and it can make predictions about a certain type of
state adopting the policy in a given year. Berry and Berry use this form of analysis to

examine the innovation process for state lotteries. Their study shows significance despite

49

the fact that only a few states adopt some policies (in the case of lotteries it was only 3%)
(Berry and Berry, 1990). It has also held up in more recent empirical applications on
such things as school choice policy (Mintrom, 1997) and anti-smoking laws (Shipan and

Volden, 2004).

Number
of States
Adopting
Policy X

 

Time

 

Figure 2.3: Diffusion Curve Model

While the EHA methodology now seems to be a standard one in this type of
study, there are several other findings and theories of diffusion and innovation that have
been proven and remain untested using this newer methodology. Glick and Hays argue
that reinvention of policy and extent of adoption (differentiation of deep and superficial
adoption) are highly relevant to the understanding of policy innovation, but are not
reflected in current models (Glick and Hays, 1991). They demonstrate that a state
adopting a policy later, may look less innovative than one adopting early because by time
the late state adopts the policy it has be reinvented and is more innovative than in its

earliest conception (Glick and Hays, 1991). They found that the early adopters (of living

50

will laws) have less facilitative laws, but that the correlation between early adoption and
facility is not strong but is in expected direction (p. 843). Perhaps their major conclusion
and contribution to the field is the finding that policy reinvention occurs along a
“common dimension”, but that particular provisions lead to reinvention in more than one
direction (p. 847). As such, it is unclear how to treat states that adopt immediately after
the first one.

There is a wealth of evidence in the fields of welfare policy and education policy
that demonstrate the context of state policy domains in which ECE must operate.
Welfare policy, a big target of the devolution revolution of the 19808, allows for testing
of the role of federalism in state policy. Lieberrnann and Shaw found is that national, not
state factors are most important in shaping state policies for welfare (Liebermann and
Shaw, 2000). Peterson and Rom found that in the absence of federal regulation, there is
wide variance across states in how they set their benefits levels and how they respond to
increases in the poverty rate (Peterson and Rom, 1989). S088 et al. found that when the
federal government enacted major welfare reform in the form of TANF legislation, states
were simultaneously forced to respond to a reduction in federalism. They found that
political and social characteristics of the state were most determinative of how stringent a
state’s policy response would be (Soss, et al, 2000). Wong tested the “parity to
dominance” theory of state financial support for education and found that in fact only a
few states actually moved flom parity to dominance (Wong, 1989). Finally, Welch and
Thompson argue and produce support for the view that “incentives provided by the
federal government do stimulate the diffusion of policies through the states” (Welch and

Thompson, 1980). They found that federal affected policies do have significantly faster

51

rate of diffusion than state preserve policies and that positive fiscal impact policies have a
faster rate (than negative or indirect impact policies) (Ibid, 1980).
Theory and ECE Policy

Based on the above summaries, and on the presumption that these theories are
particularly suited to address any policy, it is now appropriate to turn to how this all
applies to an examination of ECE policy. One attribute of ECE Policy is that over time it
does appear to have areas of both incremental change (Head Start budgets go up
marginally each year) and major change (large increases in state preschool funding in the
20008). As such, MS theory offers structures through which to understand this policy
area. Furthermore, their specific elements of why incremental change happens when it
does and why major change happens when it does, promise to shed light on the
development of ECE policy.

An additional attribute of ECE Policy is issue flaming and agenda setting. As
ECE policy has been developed over time, different problem definitions have set this
policy domain on widely different courses. Via flames, policymakers have limited the
options available for use at different times in the history of ECE policy in order to get the
issue on the agenda and enact policy. Yet the alternative policy solutions resulting flom
this flaming often has led to watered-down policy that strays far flom the original
intentions of experts and politicians.

Finally, diffirsion policy presents a framework for examining the spread of ECE
through the states — the current hotbed of policy. This is an invaluable quantitative

presentation of how different attributes of states, national policies and flaming affect

52

policy outcomes in this issue area. An EHA of ECE policy holds a great deal of promise
for understanding why states adopt policy.

In the following three chapters of the paper, I apply these theories with as much
fidelity as possible to ECE policy. I examine problem definition (and attendant solution
alternatives) for what I designate as the three eras of ECE policy. I examine flaming of
ECE policy by comparing it to flames for welfare policy and education policy — the two
policy parents of ECE. Finally, I perform an EHA analysis for ECE policy in the 50
states shedding light on what state factors affect ECE policy adoption and how

reinvention occurs among adopting states.

53

Chapter Three: Problem Definitions in ECE Policy

The current realities of ECE policy seem to present a mixed message: There is
growth in this policy direction. But there is also stasis. There is consensus and political
will, but there are also competing policy domains. In order to begin to clarify the
attendant issues of ECE, I begin with the “problem definition”. Based on Kingdon’s
theory of multiple streams being dependent on problem definitions, flaming of an issue
and attaching solutions to problems (or vice versa), it is clear that while policy evolves
incrementally at times, there are times when a burst of activity results in more major
change. The problem definition — that is how key players in a given policy domain
capture the current public, political and/or academic definition of what problem in society
they are addressing — is key to the outcome in policymaking. As such, this is where I
begin. In order to do this I present a brief history of ECE in the United States paying
particular attention to a)the definitions of the “problem” at different times b)the concepts
of who such policies serve and c) the goal and role of government action in ECE. This is
certainly not the only way to arrange this information. However, it is my primary goal to
highlight the competing and sometimes conflicting definitions and goals in this policy
arena. And by going back to the origins of this policy arena, I hope to better illustrate the
notion that the variation in ECE policy that we see today has deep political, sociological
and historical roots that directly affect the policies developed and enacted today.

I separate the history of ECE policy into three general eras. The general dates of
each era are selected to reflect times when major shifts seem to have occurred in this

policy area. In this policy area, there are the pre-Head Start Era, the Head Start Era and

54

the post-Head Start or Modern Era. As such, I divide my analysis of ECE problem

definition this way.

 

Pre—Head Start Head Start Era Post—Head Start
18008 1950 1970 1980

Figure 3.1: Three Eras of ECE Policy Timeline

It is clear to see that these “eras” are not of equal time; however, I selected them
precisely because they represent different eras of problem definition and policy
formulation. The choice to present a “post-head start” era is likely to be challenged. It is
defensible, however, because beginning in the late 19708 HS had failed to fulfill the steep
promises made on its behalf to serve poor kids and ECE policy shifted away flom federal
action to the states (V inovskis, 2002; Kirp, 2007). Beginning in the 19808, many state
governments escalated their efforts at preschool policy enactment and federal child care
policies were developed sending money to the states for welfare families at funding
levels rivaling those of Head Start. It is clear that at some point, Head Start was no
longer the dominant horse in the race for ECE policy. In fact, in 2004, the number of
children in state preschools eclipsed the number being served by Head Start.

I present the problem definitions for each era by first presenting the policies of
that era and then analyzing the problems policymakers and academics of each time was
concerned with solving. I focus on a few of Kingdon’s concepts of policymaking namely
that problems are defined either by statistics revealing a concern, a dramatic event or
feedback from an existing policy. Policy solutions then are analyzed for either having

technical feasibility or value acceptability

55

Pre Head Sfitart ErrrL’olicy

There were four ECE policies publicly enacted prior to Head Start (1965). They
are related only in that they were attempts at ECE policy prior to the large program
known as Head Start. The first two took place in the 18008: the infant school movement
(beginning in the 18208 and dying out in the 18308, primarily in colonial New England)
and the kindergarten movement (the first one in Wisconsin, 1856; the first English
language one in Boston, 1861) (Fuller, 2007). The infant school movement rose and fell
quickly, but led the way for a second public attempt at ECE in the form of kindergarten in
the 18608. While kindergartens took time to gain universality (1920: 481,000 kids -)
1950: 1,000,000 kids), they were also quickly absorbed into the grammar school policy
arena, with less and less attention paid to the different needs of young children or the
different developmentally appropriate practices and teaching methods (Tyack and Cuban,
1995; Vinovskis, 1995).

The second two pre-Head Start Era policies happened in the midst of major
national economic events: the Great Depression and World War 11. Prior to the Great
Depression, day care centers and nurseries were custodial in nature and largely used by
poor working mothers (V inovskis, 2002). In the 19308 as the country fell into the
economic tailspin known as the Great Depression, the federal government funded
emergency public nursery schools via the Federal Emergency Relief Act of 1933 (F ERA)
and then via the Works Progress Administration of 1934 (WPA). By 1937, WPA
sponsored 1,900 nursery schools for 40,000 kids. By 1943, the WPA nurseries were
gone. Yet, with WWII in full swing, there was a subsequent rapid increase in the number

of working women including those with very young children (V inovskis, 2002). As such,

56

Congress passed the Lanham Act in 1941 which funded child care centers for 600,000
kids (between 1942-46).
Problem Definitions in the Pre-Head Start Era

Vinovskis paints a vivid picture of colonial American families and schooling that
demonstrates that the common goal of education generally was moral education. He
notes that, “Rather than seeing the enactment of these early school laws as evidence of
the disintegration of the stable family in New England, it is probably more accurate to see
it as a reflection of the attempts to promulgate correct religious views and to overcome
the growing indifference of many families to religion and home education” (Vinovskis,
1995, p.7). This reflects the colonial reality that educating and catechizing were done
simultaneously and were interchangeable goals of families and churches. . .and
subsequently of schools (Vinovskis, 1995). While many professed the complementary
roles of family, church and school in young children’s lives, in many ways, the school
came to be increasingly dominant - “the last bastion for civilizing youths growing up in
the cities” (Vinovskis, 1995, p.12). This highlights the problem definition of the time for
education generally: children growing up without values or literacy were a public
menace.

Given this backdrop of the colonial family and the interweaving of morality and
education, the infant school movement began with the lofty intentions of some of New
England’s high society women. The “idea” of infant schools did not originate in
America; rather it was an import idea flom England. This “idea” emerged in the US in
the opposite way as it had in industrial England (Vinovskis, 1995). While England’s

leaders were concerned with social problems and then learned how to establish and

57

manage infant schools, in the United States, William Russell (an academic and first editor
of the Journal of Education) and others took an intact English infant school system and
integrated it into the primary system already in place in the US. ...with the focus on
introducing new pedagogies, not creating new institutions (Vinovskis, 1995, p.22). Thus
the problem definition for educationalists like Russell was lack of appropriate pedagogy
and developmental practices for young children; the proposed policy solution was
separate infant schools.

In the early 18208, Infant School Societies emerged in New England. These were
dominated by women, primarily, Trinitarian churchwomen bent on saving the city flom
moral and spiritual laxity (V inovskis, 1995). Women and these same sorts of Societies
had already helped start primary schools (for as young as age four); had brought religious
instruction to city’s seafarers, founded Sunday school program for poor children (not just
on scripture, also writing and reading). Ironically, infant schools were originally set up to
combat ills of urban life, but emerged and disappeared before the major boom in
Boston’s growth (Vinovskis, 1995, p.25). In contrast to the pedagogical focus of Russell
and others importing England’s infant schools, society women focused on the link
between crime and education with the belief that crime was a growing urban problem and
education (rather than incarceration) was the way to combat it starting with the youngest
children. Without much regard for pedagogy, reformers focused on breaking the cycle of
poverty, “by reaching young children and providing them with the social norms and
necessary skills to escape flom a life of continued dependence on society” (Vinovskis,
1995, p.26) Reformers sought reform as proactive policy (prevent ills of poverty) in the

US, not a reactive one like in UK (keep kids flom life of crime).

58

It was a quick policy movement, with the first school established in 1827 and by
1835 infant societies and schools were gone:

“By the end of 1835, almost all public comment on the infant schools in

Massachusetts had ceased, and the activities of the once-flourishing infant school

societies faded—not only flom public memory, but even flom the recollections of

those who had actively participated in the movement. When kindergartens

became popular in Massachusetts in the 18608 and 18708, they were greeted as a

unique European contribution with almost no association to the earlier infant

school movement.” (V inovskis, 1995, p.18)
Ironically, though they set out to address the moral and educational needs of poor
children, the society women who funded and ran the infant schools ceased doing so not
because the problems had gone away or been solved, but because it had become en vogue
for them to be at home more with their own children since the early years were
increasingly seen as so critical. Interestingly, it was this focus on what young children
could demonstrably learn at younger and younger ages that captured the popular
imagination, which was in contrast to Russell’s intent to introduce developmentally
appropriate experiences for infants (V inovskis, 1995, p.29). Russell wanted infant school
pedagogies to influence primary schools; not for people to expect results of primary
schools at younger ages with same methods (Vinovskis, 1995, p.30). This set the stage
for a recurring theme in ECE problem definition: contrasting notions of what is possible
and appropriate for young children’s learning and environment.

Even though the infant school movement essentially ended and ‘went away’, it at
least gave some permanence to encouraging parents to send kids to school at young ages
(Vinovskis, 1995, p.19). In the 18608 and 18708, there was a resurgence of attention to

early education as Froebel’s ideas on “kindergarten” spread flom Europe to America.

There are several reasons why the kindergarten movement “stuck” and the prior infant

59

school movement one did not. One big one is that the problem definition was more clear
(it was not muddled between the society women’s aims again crime and academic’s focus
on differing pedagogy). The problem kindergarten purported to address was the fallout
of massive urbanization and industrialization for poor kids. In addition, the initial
implementation of kindergartens in the United States (privately funded and run programs
matching middle class women with poor children) meant that the program was away
flom the large entrenched bureaucracy. Two other aspects of the original kindergartens
that had more broad appeal than their policy forbearers in the infant movement were that
kindergartens took no children under three, so they were less threat to home/ family
influence, and that proponents of Froebel’s kindergartens did not emphasize the
intellectual development of young children (over other aspects). It is also notable to
mention that the flaming of this form of ECE as a “garden for children to grow in seemed
more appropriate than institutional settings and rubrics” (Vinovskis, 1995, p.42).

The scientific and philosophical underpinnings of ECE policy are important to
trace with the policy developments as they provide grounding in what information
policymakers had available to them and what solutions were swirling around in the policy
primeval soup. The understanding of young children and child development during the
infant school movement originated flom concurrent European intellectual traditions
including Rousseau, Pestalozzi, Owen and Locke (Vinovskis, 1995). Yet these
intellectual traditions were conflicting flom the outset resulting in conflicting views both
of early childhood and of the role of families and schools in these early years. While
Rousseau’s premise that education be a child’s way to be fleed flom the corruptions of

society led to the dominance of boarding schools for children, Pestalozzi viewed what

60

was done at home as equally important to anything done in school. Thus the problem
was not “home” as a crude version of Rousseau might read, but that what was being done
in both places was so important an intervention was necessary. Pedagogically,
Rousseau’s application was very naturalistic - to expose kids to natural world in a way
that they’d ask own questions and explore world in own way. However, it is important to
note that none of the applications of Rousseau emphasized institutional settings for
children under 5 (Vinovskis, 1995).

Pestalozzi’s evidence that infants could learn at an early age led to his own policy
solution that girls and women (the most likely at home attendants of children) be taught
the latest educational techniques (Vinovskis, 1995). His dual emphasis of early learning
and at home educating were quite contrary to Rousseau’s notions of removal. Of great
concern is that at this point no real attention was paid to families where parents weren’t
enlightened or fit teachers of their children (Vinovskis, 1995). In England, Robert Owen
(l816)’s New View of Society thesis advocated that government should educate starting at
infancy leading to infant schools in 1818 and 1819, “the infant schools were perceived as
a means of dealing with pressing social problems” (Vinovskis, 1995, p.21). At odds were
the notions of holistic development versus academic emphasis and the administration of
programs and services either at home versus public programs. As Froebel developed the
early kindergartens (the first one in Blankensburg in 183 7), he advocated, “creating a
distinct organization that would serve young children. He believed that all youngsters
would blossom in rather uniform ways if nurtured at home and in kindergarten according

to his pedagogical principles” (Fuller, 2007, p.36).

61

The problem definition for kindergarten reformers was not only the lack of
appropriate development for many young children, but also the lack of appropriate
teaching methods of young children already in primary graded school. Nonetheless, the
first effect of publicly funding kindergarten and including it into the public school system
was that teachers could no longer be as personally nurturing to the children and did not
have as much direct contact with the parents (Tyack and Cuban, 1995). More
significantly, there was an instant conflict between the goals and pedagogy of the
kindergarten and those of the rest of the school system (Tyack and Cuban, 1995). What
slowly evolved was the aligning of both the pedagogy and curriculum of the kindergarten
with the graded system (mostly the primary grades one through three). Thus the very
reform that was supposed to cure problems — kindergarten - became a part of the system
where these problems persisted. The problem became not just the social distress of
troubled childhood, the problem bifurcated to include also the “problem” of past policy
enactment and implementation. This masks an additional shift in problem definition in
ECE policy at this point. Now not only were policymakers attempting to craft policy to
help young children succeed, but also they were dealing with the “problem” of existing
policy and practice trumping and transforming the intended implementation of ECE
policy.

The ideas of balanced development in childhood became a fixture in ECE debates
to come -whereby the intellectual, physical and spiritual needs are viewed as all equally
important, but not all are attended to by policy. It was not so much the notion of balance
that was challenged, rather, where the development would, should and could be nurtured.

As infant schools faced pressures to prove themselves, the intellectual gains of the

62

children in the schools became the focus (Vinovskis, 1995). “It is also clear that in the
short run, physical and moral development is much more difficult to demonstrate than
intellectual (or academic) achievement” (Vinovskis, 1995, p.33). As mentioned in detail
above, this reflected a larger academic debate on the developmental appropriateness of
different activities in early childhood.

While infant schools and kindergartens had intellectual and scientific
underpinnings, the developments of WPA nurseries and the Lanham Act, primarily had
economic underpinnings. WPA nursery schools were set up essentially as job vehicles
for out-of-work teachers (V inovskis, 2002). While the WPA emphasized education, the
teachers who filled these jobs were not trained to work with young children and the
quality of these nurseries was much less than the quality in private nursery schools at the
time. As WPA nurseries attempted in name to provide education to the children (by at
least hiring teachers-though not specialists in early childhood), the Lanham Act had an
entirely custodial view of public provision of services for young children. In order for
women to work to support the American efforts in WWII, the Lanham Act funded child
care centers for their non-school age children. Yet, many women still favored family,
fiiend and neighbor care (FFN) choosing it over the publicly provided centers
(Vinovskis, 1995).

The problem in the Great Depression was lack of jobs and WPA nursery schools
were seen as a job opportunity for out of work teachers. Any education of the children in
their care seems to have been subsequent to the economic policy aims. Likewise, the
Lanham act saw child care provision as a means to support women’s work efforts in the

war. As such, problem during WWII was not enough workers to support the economic

63

demands of the war and Lanham Act child care centers were a means to getting women
fill this void. The primary policy solution was custodial care with little mention of
education or developmental needs of the children receiving it. One unintended outcome
of the Lanham Act was a post war mentality that it was okay for women to work and
“good” child care centers were beneficial for kids (Vinovskis, 2002).
Children and Teachers in Pre-Head Start Era

The Pre-Head Start Era saw some variation in the number, type and ages of
children served by ECE policy. The two dominant “groups” of children served by policy
were poor children and children with working mothers (arguably these groups were often
interchangeable in colonial and depression times but not necessarily in WWII Lanham
centers). The children served by the infant school movement in colonial America were
primarily poor children between eighteen months and school-age (generally five or six).
(Vinovskis, 1995). The children served by charity kindergartens were initially poor
children around age five or six, but this policy quickly came to encompass all children
this age as kindergartens came to be a permanent part of the graded school system (Tyack
and Cuban, 1995). The WPA nursery schools served poor children not of school age
while their richer peers attended private nurseries in small, but increasing numbers. The
Lanham Act included all children under school age whose mothers were working in
wartime industries. Notably, many companies still flowned upon women with children
under age two working, so children this young these were less likely to be at such centers
(V inovskis, 2002).

From these inauspicious origins, women have predominated ECE teaching —

particularly white, middle-class women. Teachers and caregivers are a major element of

64

ECE policy. Women were seen as suited to running infant education and women were
leading patrons (V inovskis, 1995). There was little attention to training of teachers
though debate centered on making such teachers (and the curricula) different flom school
teachers. Kindergarten policy faced some of the same concerns and focused on child-
centered, nurturing settings with less initial attention to academics. As such, teachers
were valued for nurturing, mothering and attention to health and manners (Tyack and
Cuban, 1995). WPA nurseries were created to keep previously working teachers
employed. . .with little attention to the fact that these elementary teachers were not trained
or equipped to manage nursery classrooms filled with young children (V inovskis, 2002;
Fuller, 2007). Lanham Act centers were primarily custodial in nature and while some
may have had academic components, this was not essential to, expected of or intended by
the policy itself.
Government Role in E CE in Pre-Head Start Era

The public role in early childhood was evolving and changing during colonial
times. While the church was certainly seen as an appropriate place for moral education,
its links to literacy and other learning became strained as fewer went to church and such
activities were taken on by public schools. In addition, there was a strain between the
advocates of the movement and those already entrenched in primary education
(Vinovskis, 1995). Russell, who is credited with importing the policy “idea” flom
England, wanted infant school pedagogies to influence primary schools; not for people to
expect results of primary schools at younger ages with same method8(Vinovskis, 1995,
p.30). In addition, primary schools and school boards were not sold on infant schools,

citing “evidence’ that primary teachers preferred students without training in infant

65

schools (Vinovskis, 1995, p.30). Likewise, there were differing views of how schools
should be funded. Officials who were not spending much on primary schools were even
less inclined to spend on even more expensive infant schools.

These conflicts over the institutional setting and funding of infant schools
paralleled the growing concern about the appropriateness of educating infants in
institutions in the first place. This foreshadows something critical in ECE policy: even
after a program/policy is established, two ongoing debates rage on — the details of the
program are debated and the very existence of it is challenged. Pestalozzi’s ideas of
infant development and home care came to America heightening the opposition to infant
schools. “Both the theorists and the sponsors of infant schools belied their uneasiness in
perpetuating the unnatural act of taking children flom their homes when they defended
the infant schools as homes themselves—homes likely to promote the happiness of both
children and society in the ways that natural homes would never do” (V inovskis, 1995,
p.31). This tension between the natural “home” setting for young children and the
unnatural “institutional school” setting for young children was not to go away. It was
recognized that if most learning happened at home poor children were stuck in a losing
cycle. Yet, the focus of the leading advocates — Boston society women — turned to their
own efforts at child rearing as they quickly concluded that infants were better off at home
with informed mothers [like themselves] (V inovskis, 1995, p.32). As the funding,
administration and purpose of infant schools were debated, the role of the public in early
childhood was slowly inserted. A seed was planted that while it was debatable what role
the government should have it was less of a concern than that the government had a role

at all.

66

At the onset, kindergarten like its infant school predecessor was meant to be very
different than graded grammar school. As Elizabeth Peabody articulated it, “this human
scale-institution would not resemble a school but was a very different kind of setting”
(Fuller, 2007, p. 39). However, as kindergarten grew and was more prevalent across the
country, it was “absorbed into the public schools, its liberal-humanist ideals and the
creative practices devised by Froebel and his descendants began to fade” (Fuller, 2007,
p.39). Peabody spoke out against public takeovers of kindergartens (Fuller, 2007, p.39).
This absorption was all-encompassing, effecting the materials presented to children, the
teachers assigned to them and the expectations set for them.

“Kindergarten classes were sucked into streamlined school systems during the

first half of the century, becoming yet another grade level, plugging into

elementary school curricula and staffed by teachers adorned with higher
credentials. Teachers’ home visits and warm relationships with parents gave way

to professionals encased in classrooms, just like real teachers” (Fuller, 2007,

p.40).

In fact, there is long a history of reform efforts in schools marked by slow and
incremental impact on practice and learning. Like many preschool proponents today, the
early proponents and founders of kindergartens in America saw public schools as a
different type of schooling with a different type of pedagogy (Tyack and Cuban, 1995).
Altemately, they thought kindergarten could be “a cure for the urban social evils as well
as a model of education for young children” (Tyack and Cuban, 1995, p. 65). Activists
saw kindergarten as a place where social ills and poverty could be cured and mitigated.
In addition, policy makers thought that the child centered pedagogy and soft, open-ended
curriculum of kindergarten would positively influence the graded public system

For WPA nurseries, the govemment’s role in setting this ECE policy was limited

to providing jobs to displaced elementary school teachers — not unlike the focus of Head

67

Start on providing jobs to family and neighbors of enrolled children. This established a
government precedent for assuming elementary teachers could simply be plugged into
preschool classrooms without need for other training or certification. While
policymakers at the time acknowledged a shortage of preschool teachers, this was not met
with concern or additional policy to promote a credential; it was cast aside in lieu of job
provision during economic hardship (Vinovskis, 1995). The Lanham Act centers made
indelible the direct link between working mothers and ECE policy. It put forth a (weak)
federal policy solution: custodial care for children of working mothers until the war’s
end. In other words the federal role in child care at this point was very limited and was
based primarily on the economic needs of the country rather than the development needs
of the children in care.
Conclusions on the Pre-Head Start Era

The origins of ECE policy set the stage for what was to come. The efforts of the
infant school societies in 18208 New England were short lived, but demonstrate that flom
the beginning there were strains to establish early childhood as child-centered versus
didactic, and as part of primary schools versus separate entities entirely. The
kindergarten movement was more successful in that it lasted and kindergartens eventually
ceased serving only poor children and became universal. However, the attempts to keep
it separate flom graded schools were thwarted as were original designs of home-visits,
and play-centered developmental programs. The history of kindergarten demonstrates
that targeted programs in early childhood, in order to appeal to the masses to gain
support, are often co-opted both by the K-12 system and by the middle and upper classes

realizing and seeking the professed benefits of a given policy for their own children. The

68

WPA nurseries and the Lanham child care centers demonstrate something entirely
different about ECE policy — that in order to serve other policy and political ends (in this
case economics) —— the federal government has been able to step in and create policy flom
scratch. However, it is a cautionary tale as well, as in both cases quality was set aside in
favor of achieving high numbers served (teachers employed in WPA and children of
working women in care in Lanham). In addition, both of these policies were set up as
temporary solutions to economic problems, not as long-standing federal involvements in
either case.

In terms of problem definition, the pre-Head Start era shows how differing
definitions and changing problems lead to disagreements about the policy intent and to
varying outcomes. The infant schools were set up. to address essentially two problems:
the consequences of poverty for young children and the pedagogical inappropriateness of
primary schools for early learning. This was a hint of subsequent challenges in this
policy area as the “problem” policymakers were attempting to address often was as much
the past policy shortcomings as the actual problems themselves. While kindergarten
policy was longer lasting, it too saw the problem definition morph flom the problem of
poor children being ill-prepared for school and life (lack of hygiene, manners and
literacy) to the problem of kindergartens being just one of another grade in schooling.

The WPA nurseries and Lanham Act centers were set up not to address early
childhood or early learning needs, but rather to serve economic needs of adults. Thus the
problem these ECE policies were attempting to solve were temporary economic concerns
of working adults and the national economy. Compared to the child-centered problems

infant schools and kindergartens were focused on, these seem ill-placed in this policy

69

arena. However, the long view of history allows us to see that this is not unique to this
time period. Both Head Start and the Child Care-Welfare funding of the 19908 included
adult employment as goals. This highlights something crucial about ECE policy: that it is
interwoven into family policy and that the problems poor children face are also problems
their parents and neighbors face. In ECE policy some have found a way to produce a
single policy solution to meet both needs. The results are as confounding as the notion
that jobs created for poor adults can double as quality programs for poor children.

Head Start Era Policy

The 19608 ushered in a whole new era for ECE policy. Compared to the
aforementioned pre-Head Start era, this decade saw both a dramatic shift in domestic
policy at-large and permanence to the issues and challenges of ECE policy in policy and
public consciousness. Beginning with J FK’s Administration, major changes were
underway in education and social welfare policy. While J FK’s attempts at major federal
involvement in these areas were ultimately thwarted in lieu of other concerns, legislation
and policy set under his tenure did pave the way for LBJ ’s massive domestic agenda - the
Great Society Programs (Vinovskis, 2002).

A8 LBJ took office following Kennedy’s tragic assassination, the stage and
urgency was set for action on poverty and inequality. As the research and theories of
Piaget, Hunt and Bloom, Gesell and others brought science and philosophy together
demonstrating the critical importance of learning and development in the early years,
attention to policy for early childhood expanded at rapid speed. Because of the broader
movements in domestic policy, ECE policy was crafted not only in light of research

findings, but also in light of political expediencies and realities of the times. Evidence of

70

the disparity of mental, health and education of poor families was featured in a series of
commissions, taskforces and legislative committee hearings. Ultimately, the Economic
Opportunity Act was enacted in 1964 establishing a massive set of programs aimed at
wimiing the “War on Poverty”. These programs were largely administered by the newly
establishing Office of Economic Opportunity (OEO) led by Sergeant Shriver and were
implemented and run by local entities known as Community Action Programs (CAPS).
While CAPS were allowed to and encouraged to include early childhood programs in
their poverty programs, it was not required. By the time Project Head Start stared in the
summer of 1965, two significant realities were already in place: the Elementary and
Secondary Education Act was enacted (April 11, 1965) and tension between OEO and
local CAPs was already brewing. Interestingly, an estimated 60 percent of CAPs had in
fact elected to include ECE in WOP programs by time Head Start began (Vinovskis,
2002)

The first summer of HS saw more than 560,000 children served at a cost of $84
million ($150 per child for eight weeks). Most of the initial funding went to daily
operations (75%) with the remaining funding going to medical, dental and psychological
care, nutrition, parent recruitment, research and administration. Most of the children
were age four, five or six (though this is older than the current population of children
served). In addition more than 100,000 jobs were created by this program including
60,000 neighborhood and parent workers (Vinovskis, 2002). It quickly grew into a year—
round program as well. Yet early design choices made in the haste of political
expediency seem to have had long-lasting and often negative effects. The choices to

severely under-price the cost per child, to compromise teacher quality by ignoring the

71

shortage of qualified applicants and to hire non-trained neighborhood workers and
parents of participants remain issues for policymakers today — forty years later. Because
of the pressure to demonstrate success and keep the program on a grand scale, the
professed “impact” of Head Start was overstated initially and furthermore, it made
subsequent policy changes nearly impossible as no one wanted to tweak a successful
program or spend more money on teachers, when the program was already supposedly
working.

Subsequent to Head Start (almost immediately actually), policymakers and
researchers saw the fade-out effects of Head Start once its students reached elementary
school and set out to establish the Follow-Through program. Follow-Through was
started as both well-intentioned program aimed to sustain the effects of Head Start for its
participants and as political quick-speak to address initial shortcomings of a popular and
well-funded program. It had initial support in terms of funding, research and political
will, however, it was proposed during the changeover flom LBJ to Nixon and during the
escalation of the Vietnam War. Both factors proved detrimental to its success, however,
it remained on the books and in a select number of schools in part because of its ties with
the popular HS program, and because over time researchers and schools simply got better
at implementing intervention programs in schools. In 1968, President Nixon was not
prepared to expand ECE policy any further via Follow Through or any other program;
instead, a status quo settled into the federal efforts at ECE.

Problem Definitions in the Head Start Era
The 19508 and 19608 in America were times of great social conflict and change.

Perhaps the most dominant domestic issues were race and poverty, each seeming to boil

72

up into the wider public consciousness simultaneously, with varying policy results. The
problem that ECE policy was trying to solve during this era, therefore, is quite
complicated. That is the point: ECE policy via Head Start was wrapped up with the
broader and deeper issues of racial inequality and rampant, cyclical poverty particularly
in inner cities. Tracing these larger issues is not just a means of presenting the context of
policy in this era; it defines social issues of this era. Head Start is in name an ECE
policy, however, due to its origins in these tumultuous times, it is also meant to be a
vehicle for racial equality and an end to poverty. Not surprisingly, subsequent policy,
like the Follow-Through program and expansion of Head Start, were attempts not just to
continue what Head Start had started, but also to fix the “problem” of insufficient and
ineffective policy. It is evident that prior to many of the policies of this time, white,
middle class Americans thought mistakenly that a)blacks were doing better overtime and
that b)poverty did not exist in such an affluent society (See Moynihan, 1965; Harrington,
1962). Once these illusions were shattered, policies that ensued faced the tall task both of
articulating the “problems” rampant in society and presenting solutions that not only
solved the problems, but also appeared to do so quickly and authoritatively.

Race. Racial inequality was certainly at the foreflont of policymaking in the
19508 and 19608. Almost a century after the end of the Civil War that resulted in
fleedom for southern blacks, equal standing for them in society and the economy was far
flom a reality. As articulated by Daniel Patrick Moynihan, ‘The most difficult fact for
white Americans to understand is that in these terms the circumstances of the Negro
American community in recent years has probably been getting worse, not better.”

(Moynihan, 1965, p.2). While technically, there was equal opportunity for blacks in most

73

parts of the country, this belied two facts: a) underlying realities in black communities
prevented opportunity flom becoming reality and b) equal opportunity did not translate
into equal results (Moynihan, 1965). Moynihan brought to light the notion that blacks as
a group were at a disadvantage, and combined with racism, this meant the gap between
white majority and black minority was widening not shrinking. Moynihan placed the
blame on this squarely in the family structure of blacks at the time. For Moynihan the
problem was clear, the instability of black families at the time (whatever the cause)
trumped all other attempts at defining the problem policymakers were attempting to
solve.

There were several major political events that occurred prior to the establishment
of Head Start. A8 blacks around the country mobilized themselves in the Civil Rights
Movement, both the Kennedy and the Johnson Administrations committed themselves to
the cause of equality (Moynihan, 1965). The 1964 election (of LBJ) was indeed a
referendum on this issue. Subsequent legislation proved to be earth-shattering: both the
Economic Opportunity Act (in which HS was piloted) and the Civil Rights Act were
enacted to abolish poverty and discrimination. In addition, the landmark Brown v. Board
of Education case was the result of increasingly activist courts working to remedy past
ill-treatment of blacks particularly in schools. The problem these policies aimed to solve
was rampant, endemic inequality on the basis of race.

Moynihan argued that outcomes must be the new focus as opportunity alone had
not been sufficient to lift up blacks as a group into the mainstream of American society.
He found evidence that there were group differences between black and white families

that were crucial to understand when crafting policy in this area. While there was a small

74

stable black middle class, the majority of black families were lower and poor class and as
such the focus of Moynihan and his policies. At the time, black marriages were more
likely to end in divorce, thus there were many single mothers. In addition, the number of
“illegitimate” births in black families was much higher and growing at a rapid rate
(Moynihan cites these stats: 1940-16 percent for blacks, 2 percent for whites; 1963-236
percent for blacks, 3.07 percent for whites)(l965). As a result, there was an increased
welfare dependency and in fact most black children were on AFDC at some point in life
(Moynihan, 1965). Moynihan traces the roots of many group differences in black and
white families in the tragic legacy of black slavery in America which led to post-slavery
urbanization, unemployment and ultimately a matriarchal family structure unsuited for
the dominant paternal structure of the white middle class12 .

Poverty. Simultaneous to the civil rights movement was the so called “discovery”
of poverty in America. It was not that there had never been poverty in the US, certainly
as recently as the 1930’s Great Depression, much of the country was unemployed and

destitute. However, the 19508 and 19608 was a time of great stability and more

 

12 He notes that American slavery was the worst in the world, and that Jim Crow laws during
Reconstruction were more devastating to black males who were made dependent and separated arbitrarily
flom their families. Since the strong male father figure was the middle class standard of the time, this made
family structures of the two groups fundamentally different (Moynihan). Built upon this legacy of slavery
is the urbanization of black which was fast and flustrating-exposing persistent racism and the devastating
effects of hopelessness among blacks in urban areas. Urbanization also exacerbated the problem of
unemployment and therefore poverty. Unemployment led black farrrilies into a tailspin of devastating
consequences: exhaustion of credit, entry of wife into workforce, day-to-day existence, welfare
dependence, marriage dissolution (Bakke’s 6 stages of adjustment of family to unemployment, via
Moynihan). Even in the face of employment, the minimum wage was based on an individual’s expenses,
so for black families, women had to work as well, removing them further flom the middle class standard of
the day (Moynihan). This all coupled with the cycle of black poverty and entrenchrnent: blacks have many
kids at a young age, therefore, don’t finish school, therefore, achieve only low income level, therefore,
deprive children of opportunities, therefore the cycle repeats (Moynihan). The primary structure of the
black family is matriarchal which is problematic because it was not the mode at the time-so leadership and
power positions were even harder for blacks to attain. Subsequent results on children are tragic as well,
with higher delinquency rates, more arrests, more incarceration and the subsequent effects of such crime
(Moynihan).

75

importantly, the rise and stabilization of the heralded American middle class. As
prosperity in America led to major economic changes for much of society, there was a
subsequent presumption that this prosperity spread to all sectors and all peoples in the
American society. The discovery of poverty was simply the recognition, as highlighted
by academics, grassroots activists and a growing number of politicians, that many in
America were devastatingly poor and falling behind. Harrington articulated this
phenomenon as such, “Even when money finally trickles down, even when a school is
built in a poor neighborhood, for instance, the poor are still deprived” (Harrington, 1962,
p.9). He noted that the poor at the time missed gains of the 19308 as “millions of people
proved immune to progress” (Harrington, 1962). Furthermore, he noted quite presciently
that the “new poverty” was not temporary, but permanent, cultural and without aspiration
[prescient because the policies developed to combat such poverty naively, but garnely
aimed to eliminate poverty within ten years]. Remarkably, his characterizations of
poverty amidst a country of vast wealth hold true even today: the notion of the culture of
poverty, the idea that it is not a temporary status for many and that it is precisely the
progress America lauds and craves that causes at least some element of society to fall into
the grips of extreme poverty. For Harrington, the major problem of the day was not just
the idea of poverty as an “issue” but the status of poverty as a state of being.
Harrington’s presentation of poverty highlights two realities that were previously
unheard and likely unbelievable. First, that the poor were “upside-down in the economy”
such that societal progress actually meant worse jobs, food and shelter (Harrington, 1962,
p.12). Second, that poverty is as much an attitude as an economic status. Harrington was

particularly attentive to the different “types” of poverty: rural (as in Appalachia), urban

76

(black ghettos), drunken/drug induced poverty (flom all classes) and elderly poverty.
Harrington was before his time in noting the particular challenge of the “new-poor”,
those who worked in jobs at the fringe of old and new economies that today might be
classified as “working poor”. Harrington highlights the fact that while not all poor are
mentally ill, nor are all mentally ill poor, these two conditions are certainly intertwined as
neuroses and psychoses (sic) are more prevalent in poor people. In addition, he
highlights the research of his day that noted not only that there were risk factors for
mental health, but that they have a cumulative effect (Harrington, 1962).

An essential element to Harrington’s argument is the notion of the “culture of the
poor”. Such things as feelings, emotions and values are different because they are
formed by their harsh experience. Harrington argued that there were “two Americas” the
affluent and the poor. And how the two perspectives they represented ——a distorted
economy vs. beyond history and progress ultimately arrived at the same policy goal:
obliterating poverty. The paradox presents itself, however, in that poverty is not as
deadly in the United States as elsewhere in the world, so there is both indifference and
blindness to the hardships of American poor. In the US. poverty forms a culture and
there is an embedded interrelatedness of the different subcultures of the poor. For
example, if you start poor, you are highly likely to end poor. Harrington also pointed out
that policies in the 19608 that promised a quick and definite “end” to poverty were
misguided and destructive. Primarily because “case” poverty was misunderstood to be
temporary, not chronic. Harrington advocated a comprehensive attack on the problem of
poverty including not only programmatic aspects (social security and health care), but

also change of attitudes in the US and creation of a shared sense of purpose. He argued

77

that the federal government was the only entity capable of such an endeavor, because
cities were too damaged by urban flight and states were prone to political peculiarities.
Furthermore, private agencies were not capable of something at such a large scale. He
argued also that it should be implemented as locally as possible and that it was a matter
of political will, not improved information.

Educational Outcomes. While racial inequality and poverty were the major
problems policymakers looked to solve in the 19608, more and more attention was paid to
schools as both locus for change and as a source of the problem. In particular, the
findings of the Coleman Report in 1966 demonstrated not only that most schools were
still highly segregated (despite Brown v. Board’s landmark decisions more than ten years
prior), but also student achievement was linked more to parental background than to
schools. As such, schools were seen simultaneously as both problem (perpetuating racial
and economic inequality), and also as solution (a place to mitigate inequalities at home).
This led to two critical and competing policy solutions: increased integration and focus
on student outcomes (not just equal inputs into schools).

The Coleman Report in 1966 served as the major catalyst at the time for attention
to differences in achievement by groups of children. Amid the revelations of Hunt and
Bloom that IQ was not fixed, was the earth shattering revelation by the Coleman Report
that family and SES factors were more determinative of student achievement than were
schools. More specifically, Coleman et al, found that not only do black children enter
school at a disadvantage (flom their white peers), but they never catch up, even when
school factors are held constant (Coleman, et al., 1966). These findings were not the

expectations of the policymakers or the researchers who held vigilant the notion that with

78

the open access afforded by the civil rights movements and integration would come
increased achievement for black children. Schools (now with truly open access and
equally opportunity) were meant to be a great equalizer of children flom all economic,
racial and social backgrounds.

The Coleman Report had been commissioned by the Civil Rights Act of 1964 to
be done by a group led by Coleman and Campbell. It was the most extensive and
comprehensive of its kind with a sample of 570,000 students, 60,000 teachers, 4,000
elementary and secondary schools at a cost of $1.5 million over two years -1965 dollars
(Viadero, 2007). It was revolutionary and crucial to the policies of that era not only
because of the shocking findings, but also because of the revolutionary use of
achievement data to confirm and measure education disparities (previous education
studies focused on inputs and input equalization). The findings were so shocking and
disturbing to many of the authors and advisors that it took two years for the Advisory
committee to sign off on it (Viadero, 2007). It was written essentially simultaneous to
the passage of the ESEA and the Moynihan Report - a fact that now causes one to wonder
what ESEA might have looked like if LBJ and Congress had had Coleman’s report in
hand prior to debate and enactment.

This shock and awe response the Coleman Report had two important effects.
First, many of the secondary findings were obscured and underreported for years. In
addition, many misinterpreted this to mean that schools do not matter. For example,
second to the family, a student’s “sense of control” affected outcomes. Gaps in education
performance between blacks and whites were 1.9 years in 6th grade, but grew to 4 years

by 12‘h grade (Coleman, 1966). And while there were certainly some resource

79

disparities, these were not as much as expected, particularly in the highly polarized and
segregated south. In addition, they found that there were teacher effects in limited forms,
for example, a teacher’s verbal ability was linked with student achievement and “good”
teachers were more important for black students than for white. The study found that on
average schools have uniform effects on kids. However, the study could not distinguish
effect of individual intervention. Unfortunately, the study results and subsequent studies
fueled both sides of the raging segregation issue and this came to dominate the focus on
the Coleman Report and its findings. Though certainly given the context of other policies
at the time an equally important and lasting focus has been on outcomes and achievement
rather than inputs. One key factor that the report did was reframe people’s thinking. If
financial resources among schools were near equal with unequal outcomes, it must be
necessary to look inside schools to affect change/achievement. It is worth noting that the
policy world still believed that the black-white achievement gap would be closed by end
of century (this parallels the belief that poverty could be eliminated).

Problem Definitions and Head Start. Amid these different major problems
defined by policy and academic prophets of the day, the political workings of first
President John F. Kennedy and then President Lyndon B. Johnson, captured these
“problems” flom the policy primeval soup and produced a set of social policies unlike
any in US. history. The trifecta of poverty, racial inequality and educational outcomes
proved to be powerful motivators for policymakers of the day to do something major, fast
and with immediate demonstrable results. It is in this context that Head Start policy must
be situated and understood. It was one of many major domestic programs attempting to

solve and mitigate these problems.

80

Early childhood education as a policy issue did not present itself immediately
onto the mid-century political scene; rather it represents a coming together of many
separate but related government and non-governmental activities. The 1960 Presidential
Election (Kennedy vs. Nixon) was not dominated by education, and featured no mention
of preschool children or preschool education. However, the education debates between
JFK and Nixon centered on the role of the federal government in education (Vinovskis,
2002). Nixon was so adamant against major federal involvement because he felt that,
“When the federal government gets the power to pay teachers, inevitably, in my opinion,
it will acquire the power to set standards and to tell the teachers what to teach” (as quoted
in Vinovskis, 2002, p.16). Ultimately, Kennedy focused on two targeted policies:
construction firnds and teacher salary increases. While JFK did not pass major education
or preschool legislation, he did pave the way for increased federal role in education, for
discussions on early education and for the Great Society programs (Vinovskis, 2002, p.
24-25)

As the government was slowly coming into the fold on early childhood education
policy, there was a great deal of activity in the private sector in the 19508 and 19608.
There were experimental preschool sites in Nashville, NY and Syracuse, and on a larger
scale there were the activities of the Ford Foundation in the Great cities School
Improvement Program. The Ford Foundation Program did not specify ECE as a required
element of city renewal programs; however it came to be a major element in participating
cities (V inovskis, 2002). Interestingly, as the Kennedy Administration focused more on
ECE, the Ford Foundation shifted focus to other programs namely juvenile delinquency

programs, though “early childhood education was seen as a normal and desirable

81

component of comprehensive efforts to improve deteriorating cities” (Vinovskis, 2002, p.
30). This privately funded activity paralleled. the growing activity brewing in the federal
government.

The 19508 and 19608 also saw many changes in the views of prominent
psychologists and educationalists on child development. Until then, it was believed that
IQ was fixed at birth and hereditary-therefore analysis was on groups not individuals.
The work of Hunt and Bloom dramatically shifted the dominant view of IQ as fixed, by
demonstrating that in fact IQ could be significantly altered via a changed environment
Winovskis, 2002). Thus they argued that preschool could be used to counteract the
effects of poverty on children. Bloom demonstrated that the first four years are the most
critical for intellectual development, though neither Hunt nor Bloom at the time
advocated or pointed to one program as best because it was not yet known (Vinovskis,
2002)

LBJ picked up the momentum built by JFK in attacking the problem of “poverty”
largely by focusing on: education and training for disadvantaged youth, working adults,
coordination of federal, state and local effort; and demonstration projects (Vinovskis,
2002). LBJ saw this as a ten year program with interim and end goals (Vinovskis, 2002,
p.37). One key element of the War on Poverty and its subsequent policy incarnations
was the notion of a Community Action Programs (CAPS) approach whereby local
consortia efforts would be the ultimate arbiters of poverty progress. Most importantly
here, preschool was seen at this point (of the debates) as an often necessary starting point

of schooling to combat poverty, but it was left to the choice of local implementers.

82

Crucial to an understanding of the beginnings of ECE policy in this era is an
understanding of Education’s role in the War on Poverty - that is an understanding of the
intertwining of education policy and economic opportunity and welfare policies. When
the Office of Economic Opportunity (OEO) was created the then Secretary of HEW
Celebrezze pushed hard to separate the Education bill flom the WOP one (1964)
(Vinovskis, 2002, p.40). However, for political reasons, LBJ advisors kept the education
initiatives in the WOP bill to make it more palatable. The Economic Opportunity Act
was passed in 1964 and Shriver was appointed as the head of the new CEO. Democrats
supported the bill, Republicans largely opposed it and education’s role in the WOP in
particular was disputed. This reflected the looming concern that the problem of poverty
was one with many smaller problems to solve, one of which was education of the poor.

While the Economic Opportunity Act ultimately passed, of note are both the
Republicans’ (failed) insistence on a more prominent role for early childhood and the
importance of testimony by experts, such as Brofenbrenner, a child development expert.
He noted poverty’s debilitating effects on development in early years highlighting two
issues: the critical early years need intervention the most and poverty’s cumulative effect
(Vinovskis, 2002). What came out of Brofenbrenner’s testimony was not only the
clarification of the importance of separate ECE programs, but also the support of key
Republicans who latched onto the idea of ECE as critical in the fight against poverty
(Vinovskis, 2002). Similarly, a few key articles by prominent journalists, like Silberrnan
in the NY Times served to link (in the public’s mind) the idea that the best way to stop
poverty was via intervention in early years via developmental programs. Ultimately the

policy link between poverty programs, education and early childhood was not yet to be

83

and LBJ chose to postpone education legislation in 1964 to focus on JFK’s tax cut, the
Civil Rights Bill and Anti-Poverty legislation.

As the implementation of the Economic Opportunity Act moved along, the 1965
legislative session saw several proposals more specific to early childhood. Shriver’s
1965 proposal on the WOP was organized around four key stages in life including early
childhood, with an emphasis on prenatal care, health and school readiness (Vinovskis,
2002). Meanwhile, the Office of Education’s Office of Programs for the Disadvantaged
(OPED) proposed a large-scale summer program to run for eight weeks in the summer of
1965 serving 100,000 kids in 300 to 400 communities (V inovskis, 2002, p.72). On the
advice of family fliend, pediatrician Cooke, Shriver proposed a “project Head Start” that
would included medical screening and nutrition help and would involve parents by
requiring a quarter of staff be paraprofessionals-parents of enrolled kids.

Meanwhile, the Elementary and Secondary Education Act (ESEA) was enacted in
April of 1965 — a culmination of the work begun under Kennedy, debated in LBJ ’8 first
year and compromised into a passable federal education law. LBJ proposed five titles:
aid to poor kids (five-sixths of the money), library funding, supplementary education
centers education research labs and state departments of education. Enactrnent of the
ESEA happened quickly, but did not include provisions for children under age five. This
was despite the fact that Republicans, taking up on Brofenbrenner’s beguiling testimony
on early development, had proposed using title I for ECE. The Republican support
behind preschool was led by Representative Quie who cautioned that if preschool policy
was not enacted, any other federal aid would be in vain (V inovskis, 2002, p.84).

Ultimately, ESEA was enacted with preschool only an option for local districts receiving

84

funds, not a requirement) a few short months before Project Head Start began its initial
summer programs.

From the get- go there were talks of expansion of a program that was already
exponentially larger than any experts proposed or recommended-sealing the fate of such
things as teacher qualifications and low quality programs. The initial implementation
was entitled “Project Rush Rush” by many because the application process was so quick
and the incentive to approve many programs in order to help so many children led to
oversights of quality on the order of an 82 percent application approval rate (V inovskis,
2002). Ziegler expressed flustration flom the onset that there was a wide-range of quality
and too many poor quality ones had been funded, and proposed that poor performing
ones should be closed down (V inovskis, 2002). Unfortunately this proved hard to do and
by the 19708 this was almost unheard of. Furthermore, as the summer program surged
on, the year round program gained traction expanding flom 20,000 kids served the first
year to 218,000 by 1968 (Vinovskis, 2002).

This highlights two important issues in the problem definition of ECE policy in
this Head Start Era. First, that political need for the appearance of a massive effort at
poverty reduction meant small, well-designed and controlled demonstration size
programs would not be politically palatable. Size was indeed a required part of the
solution as poverty and the “War” on it flamed the policy environment such that small
solutions were seen as insufficient and inconsequential. Second, the link between size,
cost and quality was in many ways sealed into law in 1965 — as efforts in the decades
following to increase teacher qualifications, program quality and services offered have

achieved modest success, but have not trumped or mitigated the effect of including non-

85

qualified paraprofessionals and parents in the quality equation. Unlike size (bigger being
better), quality was not flamed as important, only as secondary.

Paralleling the strain between Shriver and researchers and OEO and CAPS, were
the differences in evaluation of CEO and HS programs. As results of the first program
came out many policymakers and researchers failed to acknowledge challenges and
shortcomings of the proclaimed success (Vinovskis, 2002). The national evaluation of
HS had design concerns in part due to the variability of HS programs and due to conflicts
over design of the evaluation and what outcome measure to use (educational ones or
not). Ultimately, OEO funded a “quick” national assessment by Westinghouse/Ohio U,
giving them less than one year to complete it. The report questioned HS’S efficacy in
making lasting academic effects, particularly in the summer program (Vinovskis, 2002).

As LBJ ’8 term came to an end and Nixon’s administration was ushered in, the
WOP and OEO were under funded (because of competing budgetary outlays for Vietnam
and an anti-tax Congress). Serious concerns about HS’s administration (in HEW or
CEO) and efficacy produced a series of new recommendations flom two prominent task
forces: the 1965 Interagency Task Force on Education and the 1966 Early Childhood
Development Task Force. The former recommended a sharp divide between the federal
role in K-12 education (age five and up to be served by HEW) and preschool (HS to be
for three’s and four’s and kept in OEO). The latter recommended that preschool/HS was
not sufficient for poor kids to succeed in school and proposed a follow-through element
in later grades in public schools (V inovskis, 2002).

The idea of a Follow-Through Program was developed modestly by LBJ who

recommended such things as special classes and field trips when HS children were in

86

elementary school. Ziegler also warned that HS was “no simple panacea for
disadvantaged children” (p.117). Furthermore that intellectual development and
remediation were not sufficient to inclusion of poor in greater society (V inovskis, 2002).
Following the initial year of summer and full-year HS in 1965-1966, the focus of policy
around the program centered on a more narrow set of issues, but never on the program
itself. Debate continued on the appropriate place to house HS within the federal
administration. Strain remained between the federal mandates for HS programs and the
local authority of CAPS to administer all WOP programs (V inovskis, 2002). In addition,
the notion of a Follow-Through program was brought to fluition as a part of amendments
to the Economic Opportunity Act of 1967.

While there was ongoing discussion surrounding the issue of teacher quality in
HS programs, the fate of this issue was sealed with a combination of professed early
success of the program as run by unqualified teachers, neighborhood caregivers and
parents. Some attempts were made to coordinate HS with policies in ESEA, however,
due to the administrative separation both at the federal and local levels this proved
fluitless at this point. The 1968 election further locked the HS debate into place for the
short-run as the focus Shifted to the Vietnam War and to a decreased federal role in
education and civil rights. Under Nixon the issues surrounding HS design, quality and
evaluation were slowly tabled in favor of a status quo. This status quo was marked by a
modified Follow-Through program, increased spending on HS and a shut-out of other
ECE issues as demonstrated when Nixon vetoed the Child Care Bill of 1971 as too

intrusive of states rights and family role in early childhood (Vinovskis, 2002).

87

Children and Teachers in Head Start Era

The children ECE policy aimed to serve in the Head Start Era were poor children
whose lives were compromised by their environment in ways that were seen to effect
their achievement in school and in life. Poverty was linked to family income and there
were really no efforts to serve other children in this era. However, increasingly other
children were indeed going to preschool and kindergarten was reaching more and more
five year olds (as mentioned in Chapter One). Of issue then was not which children to
serve by federal ECE policy, but how many could be served. As highlighted above,
many researchers advising on the issue wanted to start with a demonstration project
serving fewer children but serving them with the quality their development demanded
and the teachers trained to do 80. Yet, as HS was one effort at ending the War on
Poverty, small was not politically viable. Shriver wanted as many kids served as possible
in part to give the most the benefit and also to give the program itself more legitimacy.
As such, over a half a million children were served in the first summer - estimated to be
half of those in need (V inovskis, 2002). The compromise was made between size and
quality such that -more children served meant fewer dollars per kid which meant less pay
for teachers which meant poorer quality. While the number served dipped during the
early 19708 below the initial 500,000, the number served never dipped below 300,000
children which is exponentially higher than the initial recommendation of a quality
demonstration program for 2,000-3,000 children (Vinovskis, 2002). During this era the
issue of age of the children was not prominent, and as kindergarten was also still
becoming more universal, there was variation in the ages served by HS. Initially the

children served were much older than the three, four and five year olds served today.

88

Initially, four and five year olds dominated the programs, but due to kindergarten
enrollments, this has shifted to three and four year olds.

As mentioned already, the issue of teachers and teacher quality were prominent in
the HS Era, yet proved to be greatly expendable. Amid the implementation of Project
Head Start were ongoing discussions not only of the size and scope of the initial program,
but also the teacher ratio, the cost and the teacher qualifications. Estimates for class size
ranged flom a ratio of 308tudents:1 teacher vs. 15 studentszl teacher with 2 aides. The
proposed cost based on teacher salary was $1000/child, however the announced cost was
$180/child based on political expediency and salience (Vinovskis, 2002). The result of
this undercutting of the actual cost of quality teaching in the initial program proved to be
deadly to the long term standing of the policy: local HS applicants could not ask for more
because once the benefits (again exaggerated for political reasons) of the programs were
announced at this low cost, there was no focus of increasing teacher quality. As such,
when programs did ask for more money they did so to increase hours or numbers of kids
served, not to improve teacher quality (V inovskis, 2002).

The Republicans proved to be (as a party) more of a champion for ECE than their
progressive Democratic peers. While prominent Democrats, let by JFK and LBJ
advocated for ECE, they did so as a part of the larger agenda against poverty. In contrast,
the Republicans led by Representative Quie advocated not only for ECE as an
educational issue but also insisted on quality preschool as essential to other federal
educational interventions. For example, Quie’s amendment to ESEA (that failed)
recommended a specially trained corps of teachers and called for higher qualifications for

teachers of poor children especially (Vinovskis, 2002). “Ironically, although most

89

experts and policymakers initially saw HS as a way to compensate for or overcome the
educational and cultural limitations of the home experiences of poor children, many of
these youngsters soon were to enter HS projects taught by semiliterate parents or
neighbors rather than professionally trained. teachers” (Vinovskis, 2002, p.151)
Government Role in Head Start Era

Three elements of government involvement were present during this Head Start
Era of ECE Policy. First, the role of the federal government in education and social
welfare programs was hotly debated. Second, the relationship between the federal
government and local entities (both school districts and CAPS) was of issue. Finally, the
notion of comprehensive services and a comprehensive approach to the myriad
challenges of poverty was put in contrast to a more targeted, limited approach.

The role of the federal government in education exploded with the passage of
ESEA in1965 and has never been the same since. And while there were debates prior to
the time of enactment about whether or not the federal government had an appropriate
place in education policy, ultimately enactment of ESEA silenced this debate for at least
a decade. Likewise, the Economic Opportunity Act committed the federal government to
a major funding effort against poverty (on the order of close to $1 billion per year).
These social and economic and educational programs however were to be carried out by
CAPS — local government and community entities for whom the CEO was seen primarily
as a funder, not a regulator or rulemaker. In contrast to the initial permutation of ESEA
which saw limited federal prescriptions for implementation, the CEO was heavily

involved in what and how CAPS did their work. But as with education, the debate about

90

federal involvement ceased to center on whether it should or should not be involved and
shifted to how it should be involved (simply as funder or also as implementer).

The grass roots success of the civil rights movement helped to fuel the notion that
local community based action worked best for local communities. However, flom the
federal perspective this meant wide variability and wide swaths of quality in
programming across the country. While local CAPS were likely best at sensing and
meeting local needs, they too faced political compromise. This variation and strain
between a heavy handed federal role and a loose-oversight role, trickled over into Head
Start and ECE policy at the time. While HS was more politically popular, it suffered at
the hands of wide variation in quality and administration. In major ways the employment
function of HS was seen as outweighing the educational function for children (V inovskis,
2002). This variation flom the start meant challenges in evaluation and policy reflarning
for years to come.

A final concern about govemment involvement at the time involves the notion of
comprehensive policy and comprehensive services. Both Harrington and Moynihan
argued for policy that addressed all the attendant needs of the poor and of black families.
Harrington presciently noted that simply giving jobs to poor people was insufficient as it
ignored the endemic issues of chronic poor health care, mental illness and family
dysfunction. Likewise, Moynihan cautioned that any piecemeal approach was
insufficient and that, “A national effort is required that will give a unity of purpose to the
many activities of the Federal government in this area, directed to a new kind of national

goal: the establishment of a stable Negro13 family structure” (Moynihan, 1965, p.2).

 

13 While Moynihan used the term “Negro” which was possibly appropriate at the time, I use the term black
and black American interchangeably with his term Negro, as it is more colloquially acceptable today.

91

These paralleled the discussions among researchers and policymakers about the approach
to ECE policy - should it just be academic, or include a fill] range of developmental and
social services.

Conclusions on the Head Start Era

The 19608 were an era of major domestic and social change in the United States.
The issues of race, poverty and educational opportunity exploded simultaneously onto the
national political agenda in many ways overwhelming a system meant to limit quick,
swift, major federal policy activity. The results at the time were astonishing: major
legislation on Civil Rights, Education, Poverty and Early Childhood Education.
Policymakers and experts at the time were tasked with the daunting challenge of ending
poverty and racial inequality and fast. Emerging information flom research demonstrated
that IQ was not fixed, that environment mattered to a child’s full development and that
something could be done about it. This was a compelling area in which to focus the
solution and success of the agenda against social and racial inequality. The idea was
made colloquial that the federal government should and would fund preschools for poor
children and would be successful in helping those children escape the grasp of
impoverished lives.

While the problems that HS was professed to be solving were daunting, so too
was the task of implementing a politically popular program with the weight of the policy
world on its back. The notions of small, smart demonstration programs gave way to
large-scale programs with watered down quality provisions. The notion of solid
evaluation to determine effectiveness of the program gave way to quick shot evaluations

and political reports flaming the results as successful — at the expense of the actual

92

success of the programs. In effort to demonstrate that the problems of poverty for
children were being “solved” by Head Start, a new problem began to bubble up from
policy primeval soup: the policies themselves. The subsequent efforts at Follow-Through
to sustain supposed success in raising IQ in Head Start were not ever given a chance
because they were sustaining a myth and they were competing with a change in
administration and a massive and unpopular military involvement in Vietnam (Vinovskis,
2002)

Problem definition in the Head Start Era demonstrates two crucial things. First, in
terms of problem definition there is the sense of attaching small sellable solutions to a
massive (possibly incurable) problem. Early childhood education was just being
understood to be immensely important for all children in terms of IQ and life success.
That this happened at a time when issues of racial inequality and devastating poverty
were being attacked is consequential. This coupling of early childhood and the problems
of poverty has proved long-lasting as it is rampant in ECE policy efforts today. What it
meant for ECE policy was that the first major long-lasting attempt at ECE policy was set
on a foundation of enormous expectation and was set forever between the world of
education (where leaming and achievement are the goals) and the world of welfare and
social services (where jobs and self-sufficiency are the goals). By attempting to solve
both educational and economic concerns in early childhood policy, the Head Start
program design was somewhat stranded in the balance — it did provide programming for
preschoolers, but it did so without trained or well-paid teachers in order to fulfill the

economic needs of parents.

93

Post-Head Start Era Policy

As the focus on changing HS faded and modest tweaking and general political
glad-handing perpetuated the program into the 19808, there was a shift in ECE policy in
many palpable ways. The problems HS aimed to address did not go away and while the
policy streams running into the federal government dried up, states became a new arena
or venue for ECE policy. In addition, due to both the “success” of working women
during WWII, the feminist movement and the increase in single-parent families, what
was dubbed “child care” policy also came to the forefront of many a domestic policy
agenda. In addition, the late 19708 and early 19808 saw the K-12 education reform
movement reach a fever pitch as commission after commission chided the current ability
of American schools (and students) to educate (and compete) in an increasingly global
economic reality. These streams - state preschool policy, working women’s child care
needs and K-12 school reform - came together in various ways in this Post-Head Start
Era of ECE policy. In many ways this era reflects many of the pre-existing conflicts
and issues of the HS era, however, while the HS era tended to focus on poor children in a
targeted, albeit conflicted goals, way, the modern era has come to encompass all children,
all families, all K-12 school issues, all working family child care needs and all early
learning spectrum goals. The modern era is a glut of problem definitions and solutions,
not the least of which is the problem of past policy that so entrenched, there is little
impetus for real change.

State ECE policy began slowly during the 19708 but really accelerated during the
19808 as a part of standards-based education reform and the 19908 as a part of

accountability reforms. At the time Head Start was enacted only New York and

94

Califomia put policy in place to serve preschool children. It was not until the 19808 that
significant numbers of children began to enroll in state-run and funded preschools and
ECE programs. As of 2007, thirty-eight states have enacted at least one ECE policy and
the number of children served by state preschools now roughly equals the number served
by Head Start preschools (NIEER, 2007). Funding has escalated rapidly as well,
however, unlike HS which has seen a fairly steady increase in funds (see chapter One for
more on this), state preschool funding is less stable and has not increased in such an
incremental way. Nonetheless, state preschool is now a dominant force in the ECE
policy arena.

While Head Start policy has been tweaked during the Modern Era, most efforts to
do so have not significantly changed the design, implementation, teachers or children
served. Recent efforts under Bush II’s Administration have focused on increasing the
number (or percentage) of teachers with Bachelor’s degrees and on doing large-scale
assessments to measure achievement of participant children. Yet many of the original
design elements that are of concern, namely teacher qualifications and usage of neighbors
and parents of participating children, have remained intact. As such, other federal efforts
at ECE policy have focused on creating new programs and expanding the ideals of Head
Start to other children. Early Head Start, enacted in 1994 as a part of the reauthorization
of Head Start that year, serves children ages zero to three and models many of the same
elements of Head Start: parent involvement and education, comprehensive health
services, nutrition and family supports. An additional program, Even Start was initially
authorized in 1988 as an amendment to the ESEA legislation. Even Start picks up on the

19608 notions of serving families in poverty more intensively with literacy programming

95

for adults and children bypassing the notion that preschools alone empower families via
their children’s education.

Child care policy arose most prominently in the mid-19808 on two flonts: demand
from working women and federal welfare policies such as the CCDBG and later the
CCDF. In the mid-19808 the number of working women with children under six tipped
the scales (around 1985, the number climbed to more than 50 percent) and a few years
later the number with children under age three also tipped the scales (around 1987, the
number was more than 50 percent). (BLS, 2006; see Chapter One for full table). This
certainly fueled the private demand for custodial services for children with working
mothers on a scale that had not been previously seen in the United States. At the same
time, federal efforts to reform welfare began to focus attention on the link between work
and welfare and more finely pointed to the link between child care and working for
former welfare recipients. The Child Care Development Block Grant program enacted in
1990 allowed states to support entities in child care provision for children under age
thirteen flom poor and low income families. The revised version of the CCDBG came
about during the Welfare Reform of 1996 as the newly named Child Care Development
Fund(CCDF) and was modified to focus more on the activities of families to remain
eligible for services and vouchers. The CCDF represented an increase in funding over
the CCDBG levels, primarily to get more families off the welfare doles by providing
custodial care to their children so they could secure and retain jobs.

While these activities by states and the federal govemment aimed specifically at
early childhood, other efforts were underway that would also affect this policy area.

State education reform spurred by the discouraging findings of reports like the Nation at

96

Risk report of 1983 focused efforts on increasing standards for learning in schools and
often, though not always or consistently, this included provisions for educating young
children. The theory was that since young children entered schools unevenly prepared,
and standards were newly set for them upon kindergarten entry, early childhood
education was seen as a logical step towards achieving success via standards-based
reform. Similarly, in the 19908 many states sought further reform to add “teeth” to the
standards and expectations of prior reforms by holding schools and teachers accountable
for the teaching and learning done in schools. As before, many states included or added
early childhood provisions to their reforms in effort to flont load the achievement of
youngsters.

The Modern Era of ECE Policy has seen additional efforts in recent years
culminating in more concentrated and coordinated efforts aimed at quality and access. A
prime example is the current UPK movement which is made of a remarkable coalition of
national, state and local advocates and policymakers who target states at the cusp of
making ECE policy change and focus efforts into a single effort —universal preschool for
four year olds (Fuller, 2007; Kirp, 2007). This type of concerted national campaign is not
new to domestic policy; however, it represents a concerted effort by its members to focus
policy efforts at a very narrowly defined policy seen as the most palatable and defensible
for public attention. UPK advocates argue for and support voluntary universal preschool
for all four year olds, and have aligned with the momentum of the accountability efforts
in NCLB to focus on hi gh-quality, standards-based academic preschools as part of the

school system.

97

Problem Definition in the Post-Head Start Era

The Modern Era of ECE policy demonstrates an explosion of size and scope, but
also of problem definitions and coupling. While the policies of the Head Start Era
centered on a few major general problems (poverty and racial inequality), the policies of
the Modern Era have focused on more specific problems leading to varied program,
erratic quality provisions and generally, the establishment of a “permanent” policy
window in this area.

The 19808 saw a dramatic cultural and societal change peak and really leap onto
the policy agenda as working women with young children became a majority and their
child care needs became more and more of issue. As more women with young children
entered the workforce, dual-family needs became both a corporate and government
concern as “the perennial question of how government and employers can help to
strengthen working families and neighborhood supports” (Fuller, 2007, p. 54). Feminists
argued that women should not have to choose between family life and a profession and
urged a national network of child care centers as far back as 1966 (Fuller, 2007). While
the number of providers and types of options slowly increased to meet the increased
demand, there were concerns flom the start about quality and the child care workforce.
The problem for childcare, in contrast to most preschool policy was a need for at
minimum custodial care for children in safe enviromnents. This contrasted the problem
for concurrent preschool policy aimed at serving developmental needs of young children.

State education reforms have been particularly important to ECE policy in this
era. Standards-based reforms in the 19808 defined the problem as twofold: inadequate

expectations for American students and America falling behind in a global economy (see

98

for example, The Nation at Risk Report, 1983). This twofold definition led to a few
major policy solutions primarily done at the state and non-govemmental levels. Most
notable was the standards-based reforms whereby states developed statewide standards
systems for students. Many states included some form of ECE policy in their standards
reform, arguing that many students were ill-prepared for school once they got there, so
early childhood was crucial to any reform efforts (Fuller, 2007). In this way, however,
early childhood policy was more of an “add-on” policy meant to bring more children into
the “fold” of the K-12 system with little attention for the actual needs or pedagogies of
early childhood (Plank and Boyd, 1994). As such, the problem definition this ECE
policy was meant to solve was to supplement a larger goal of higher academic
achievement and economic participation in later grades.

State education reforms in the 19908 and into the 20008 took on a new bent with
accountability being implemented alongside standards. Accountability policy with its
heavy emphasis on measurable learning results favors standardized testing and a strong
emphasis on academic, not social or behavioral concerns. Not surprisingly, many argue
that this is ill-fit with the pedagogies of ECE, and concerns about “drill and skill”
practices for young children have been raised (Kirp, 2007). However, for the UPK
movement in particular the coupling of ECE with the momentum for accountability as
presented in the No Child Left Behind Act of 2002 (NCLB), has been of utmost
importance. A problem many see with the alignment and coupling of UPK with NCLB
is that NCLB’s accountability (which came after accountability measures in many states)
is very limited with an emphasis on alignment of prekindergarten learning with K-12

learning (Fuller, 2007). While alignment with accountability served to increase some

99

public support for UPK, preschool policy produced as a result of such a union may water
down the developmental and pedagogical concerns the UPK movement and others in
ECE strongly adhere to and protect. (Fuller, 2007, p.67). The NCLB accountability
approach to ECE policy has truly been anti-humanist in the worst sense —- it not child-
centered, it is not developmentally appropriate and it is not balanced. The problem
NCLB was established to address was a continued achievement gap among majority and
minority children and a lack of attention to the learning needs (the measurable ones) of
all children. This is a far cry flom ECE problem definitions that include the needs of the
young child and the particular concerns of children in poverty.

While there has been much activity in state education reform that has affected
ECE in the Modern Era, there has also been dramatic growth in what is called “child
care” policy, but is ultimately a form of “welfare policy”. In 1990, under the leadership
of President Bush I the Child Care Development Block Grant (CCDBG) was enacted
providing firnding to states to subsidize child care for poor and low-income parents of
young children. This was the first large-scale federal effort at child care since the
temporary Lanham Act centers during WWH; however, there were remarkably few
increases in quality provisions in this 1990 law flom the 1941 law. The CCDBG set out
primarily to assist adults in being able to secure and hold onto jobs by firnding child care
for their children. “Some UPK advocates would later claim that these huge policy steps,
though they did expand child care, failed to advance the real thing -— preschools linked to
the public schools” (Fuller, 2007, p.53). The problem definition for the CCDBG and its
later enactment the Child Care Development Fund (CCDF) was simple: poor and low-

income adults need child care for their children if they are working or looking for work.

100

Thus, focus was on establishing a funding mechanism sending federal dollars to state
coffers to be distributed to a cadre of providers meeting at best minimum health and
safety standards.

One mark of ECE policy that seems to separate it flom its policy parents is the
solid evidence that not only concurs within the field, but also aligns with hard science to
produce staggering scientific support for the need for solid early childhood experience for
all young children. It is surprising then that one “problem” that has arisen recently is a
watering down of the message of this evidence and often a blatant ignorance of its most
vital component — quality of experience and care. Fuller proposes that for example the
Perry Preschool success has become more legend than policy reality. As with the
promotion of early HS successes masking underlying criticism and concerns, the amazing
success of Perry has been not only in its measurable effects on participating children, but
also in securing for the ECE policy community a permanent link between science and
politics (Fuller, 2007). Fuller notes however that while the attention of large influential
non-govemmental groups was welcome in this policy domain, there was also a tendency
to misuse or under interpret results better left to scientists. For example, the Carnegie
Corp’s report over emphasized and misinterpreted the notion of “synapses” in young
children’s brains, urgent that more and more was better when the key is actually, not just
the number of synapses, but also the “pruning” of them for efficiency (Fuller, 2007,
p.51). He argues that, “What has seeped into the UPK debate is the utilitarian logic”
(Ibid, p.48).

Likewise, there is the challenging issue of scale-up of the model programs that

ECE policy is so tied to. While the evidence flom these long-term studies points to

101

significant short and long-term benefits of quality preschool, the application of them as
public policy has been flustratingly challenging. Quality concerns arise flom the start as
political tradeoffs are made in order to get policy enacted, thus compromising the
possible benefits to children flom the get-go. Thus, short-term findings flom evaluations
of such programs as Head Start, Even Start and some of the state preschool programs
show more modest results in the short-term with many results particularly in math fading
out quickly as children enter elementary school (see Chapter One for more information
on these evaluations and findings). As such, a new problem has arisen in academic and
evaluation circles as to how and if scale-up is possible on the scale needed to serve the
children who most need it. This strain between solid design and evaluation of ECE
programs hearkens back to the debates surround Head Start as policymakers like Shriver
sought large-scale programs to serve a maximum number of children while contemporary
academics urged initial programming on a smaller scale in order to work out kinks and
evaluate program elements correctly (Vinovskis, 2002). Yet, there is an inherent
challenge with attempts to try to rethink scale and implementation as now it is “the
folklore around this early program that now matters most -— these seemingly miraculous
effects are taken as sacred truths by many early educators, advocates, and journalists
across the country” (Fuller, 2007, p.47).

A final problem definition swirling around the ECE policy soup is that of
economic return. Based on the findings of the model program, economists have
calculated the costs to benefits ratios for these programs to staggering results. So in
addition to the evidence of major benefits to ECE participants, there are also societal

benefits, quantifiable in economic projects of costs saved on other public programs and

102

on such things as future tax revenue on increased income. As this evidence came forth,
several prominent economists unrelated to ECE quickly jumped into this policy arena.
Nobel Laureate, James Heckman and Federal Reserve Bank President Grunewald have
added levity to this economic argument. This represents a shift in this policy area as the
focus shifts flom the benefit of the child to that of society. The problem here is defined
as poor public investment; the solution is quality ECE programming to maximize public
dollar for dollar return over most any other policy solution return. One thing that is lost
in this connection between ECE and economic return is that while accountability
coupling has led to ECE policy focused narrowly on academic outcomes, these economic
studies profess the benefits of outcomes in many other areas including behavior (crime
activity) and social outcomes (family choices such as delaying childbirth).
Children and Teachers in the Post-Head Start Era

The issue of which children should be included in ECE policy has fundamentally
changed in the Post-HS era. It is somewhat a chicken-or-egg-comes-first-question, but
the reality is that 60 percent of all children under age five have some weekly non-parental
care arrangement and by age four (prekindergarten enrollment) this skyrockets more than
80 percent (NCES, 2005). In addition, between federal childcare programs and
exemptions, state run preschools and HS, many children are already benefiting flom
government policy. The question of “who” should be served by public programs is still
somewhat at issue, however, in such a modified form as to almost negate the prior notion
that public involvement be solely to help poor children. In other words, now that there
are arguments for and in six states programs to serve all children, all children are part of

the policy area whether or not they are ever enrolled in a public program. This is a

103

dramatic shift or reflarne of the “problem” in this issue area. Whereas before the problem
was primarily the inequality experienced by poor and black children, now the problem is
that all children do not receive public preschool.

The issue of teachers remains enigmatic in ECE policy. While the most solid
effects come out of programs run by highly qualified teachers, this variable has not been
isolated so convincingly as to eliminate the reality that most ECE teachers are not highly
qualified. As demonstrated in Chapter One, Head Start does not require a Bachelor’s
Degree (BA) of its lead teachers, but has only recently moved to require that a majority
of them in each state have BA or more credentials (Good Start, Grow Smart, 2004).
Admittedly, if HS is to be credited with advancing generally the cause of public ECE,
state run preschools have done a better job of advancing the cause of quality in ECE
programs. In fact, thirty-four of the forty-eight state programs require at least a BA for
lead teachers (NIEER, 2007). However, only one state — Vermont — requires a BA of
teacher aides who round out the class sizes and often function as full-teachers for all
practical purposes. Quality is still sacrificed at the behest of cost and since teachers make
up the most costly element of preschool funding, tweaks are often made to other quality
elements such as class Size, in-service training and support, and cuniculurn to
accommodate budget cuts or fimding concerns.

Government Role in Post-Head Start Era

The public role in ECE policy in this Era is solid-there is no longer much
challenge to the notion that there should be at least some public commitment to young
children. Yet interestingly, the government role is still largely that of firnder and

regulator. The federal government funds programs but these programs are run by local

104

groups and/or administered by the states. While ECE policy is primarily still a
community effort, it is even more remarkable that it is also largely a private effort. For
example, as HS developed, “the nonprofit sector has grown enormously in middle-class
communities as well, with child care organizations representing one of the largest parts of
it” (Fuller, 2007, p. 59)

As federal and state investment increases so too do opportunities for a mixed
market of provision of services. Fuller argues that vital to the success of efforts by
groups like UPK is not to presume or advocate for elementary schools to be the natural
locus of housing preschools. He suggests that since there is already a mixed market that
parents are continuing to select even when subsidies are available, public efforts should
accommodate this existing market and work to improve quality, not to decrease variation
(Fuller, 2007).

Problem Definition in ECE Policy

While young children were not an invention of any era of American history, the
policies designed for them have made a dramatic change during the democracy’s
development. What is most notable about early ECE policy is that linked with any sort of
cognitive, educational aims of policymakers were moral aims-a sure mark of the colonial
primacy of church and salvation in daily lives. As the American society evolved and
changed, the tension between institution and home particularly for placement and rearing
of young children established itself squarely in the ECE policy domain. As infant
schools fell out of favor, home environments became more publicly important. And with
the advent and swifter universalization of kindergarten came a startling demonstration of

the power of grammars of schooling and society in encompassing a sweet, albeit non-

105

saccharine policy aimed at helping poor children with hygiene, manners and ABCs, into a
larger institutional system that alienated a majority of the goals kindergarten set out to
achieve. The other pre—HS policies were essentially stop-gap policies, the WPA nurseries
to provide jobs for out of work depression-era teachers and the Lanham Act to allow
women to work during WWII while their young children were in custodial care.

The problems defined by these early ECE policies are revealing: that morality and
education are linked in family and society in very intricate ways. As such defining the
“problem” in ECE policy was tenuous flom the start: poor families were the problem-the
solution teach their children outside of the home. In addition, the issue of social class
dominated infant and kindergarten school movements with middle class values of
education and economic stability promulgating the success (and ultimately failure for
infant schools) of these policies. Interestingly, the WPA and Lanham Act programs were
done essentially as temporary solutions to larger economic problems. Thus the focus of
the pro grams was less on the need for them for children, but rather on the attendant needs
of adults-out of work teachers for the WPA policy and working mothers in the Lanham
Act policy. This highlights the direct relationship to macroeconomic forces and ECE
policy as well as the microeconomic forces of families in care for young children.

These early policies also began to demonstrate the tenuous relationship between
science and politics in the arenas of family and education policy. As Pestalozzi’s and
then Froebel’s ides of early learning and schooling spread flom Europe to America, it
became evident that while the science of early learning demonstrated the importance of
early years for later schooling and success, the politics of forming policy for young

children did not always follow science nor did science provide constant guidance.

106

Certainly the hard sciences experience such shifts as this area of social science, and
perhaps the best analogy is Galileo’s ostracism and excommunication for suggesting the
helio-centrisitic universe to a world formed around earth-centric religion and society.

For ECE policy, there is a similar tension between believers in a nurture-home based
ECE and an academic-institution based ECE. The difference is that while eventually the
universe was accepted as centering on the sun and Galileo was given his proper due, ECE

has yet to resolve these opposing views of early childhood.

107

Chapter 4: Framing Early Childhood Education Policy

The amalgam of policies across the states alone makes it clear that momentum
and consensus are misleading ways of understanding ECE policy today. It is not so
simple to say that because most kids are already in ECE programs, because most states
fund programs and because the federal government has been involved for so long, there is
clear forward movement in this issue area. Rather, a more thorough understanding of
how the existing policies have developed so differently is needed to shed light on what
factors come into play in developing ECE policy. Fortunately or not, the wide-variety of
programming across the fifty states, give us a range of policies to examine differences
and developments of ECE policies. As such, my questions are simply: why has ECE
policy developed the way it has and taken on such different forms across the states, in
particular? What arguments are made and values appealed to in the development of these
policies and to what effect?

One way of answering these questions is to look at policies that have been enacted
and highlight the arguments, facts, values and justifications made prior to and during
enactment. This involves a methodology called flame analysis — an examination of the
flames used by policymakers in moving a social issue forward to a given point. Frame
analysis comes flom issue framing theory that posits that how an issue is framed —what
facts and values are considered as “real” and valid — directly impacts the outcome of a
social movement or policy controversy resolution (see for example, Schon and Rein,
1992). Frames are used by policymakers, advocates, and the media to limit the debates
and outcomes in particular directions — and for a given policy controversy there may be

opposing flames such that the only resolution will come through the various processes of

108

flame alignment, flame transformation, flame extension and amplification (Snow, et a1,
2002). In this chapter, I present the results of a case study on issue flaming in ECE. I
utilize flaming theory to examine how governors in two states — Oklahoma and Michigan
— have used flames to present ECE as a policy and how these flames compare to those of
education and welfare, the two policy parents of ECE.

Issue Framing and Frame Analysis

In trying to link research, policy and practice, educational researchers undertake a
wide variety of methodologies to examine teaching and learning and then try to
“translate” these findings into policy that will promote improved practice. Education
policy employs a broad set of tools to bring about intended change and to anticipate and
limit unintended consequences (McDonnell and Elmore, 1986). One such tool, issue
framing, has been largely unexamined in education. Issue flames, are the underlying
structures of belief, perception and appreciation that form the foundation of policy
positions (Schon and Rein, 1992). In other words, issue flames provide people with
boundaries of values within which they should consider and act on a policy.

Frames involve both policy research and policy design, “two different but
potentially complementary traditions of inquiry: the study of the practical work through
which policies are designed in action, and the study of the flames that underlie policy
controversies - the assumptional structures held by participants in the forums of policy
discourse and by actors in policy-making arenas” (Schon and Rein, 1994, p. viii).
Because flames involve both research and design, theories of flaming can address the gap

between policy practitioners and researchers of policy.

109

Issue flaming is not a new concept in social science research; rather it has been
developed in the field of sociology and more recently adapted to empirical tests in
political science. The seminal works of Gamson, Gofflnan, Snow and Benford, Schbn
and Rein, focused on the “collective action flames” examining the flames that form and
guide social movements. To date much of the work done in this field has focused on
three connected but distinct lines of inquiry: What are the flames used in a given issue
area? (Gamson and Modigliani, 1989; Davies, 1997; Gamson, 1992; Benford and Snow,
2000). What are the effects of flaming either in discussion (Gamson, 1992; Cobum,
2006) or on public opinion (Nelson, et al, 1997; J acoby, 2000)? And, what are the
theoretical and conceptual markers of frames (Gofflnan, 1974; Gamson, 1992; and
Benford and Snow, 2000).

Some concepts and theories have evolved and been establish in the flaming
literature that are defined here for sake of clarifying some of the terminology used in this
research. An issue frame is a set of ideas and values used to understand an issue
(Gamson, 1992; Scth and Rein, 1994; J acoby, 2000). Framing is the use of frames to
move policy/society forward on an issue (Benford and Snow, 2000; Davies, 1997).
Framing effects are the effect of flames on policy outcomes, implementation and public
opinion (Iyengar, 1996; Nelson et al., 1997). Frame analysis is the retrospective study of
flames in an issue area with positive conclusions (Gofflnan, 1974; Gamson, 1992).
Framing policy is the use of flames as a tool to get a policy enacted and implemented
with normative implications (Schbn and Rein, 1994; Coburn, 2006). Tables 4.13 and

4.14 in the Appendix summarize the mechanisms as presented in the flaming literature.

110

Frames have quite particular and specific attributes that distinguish them flom
such things as fabrication, political persuasion, “spin” and mere rhetoric. Frames are not
merely semantic, nor are they merely words or matters of the mind (how one thinks about
an issue), but also “affect the way an activity is organized especially for its main social
agents” (Goffinan, 1974, p.247). This is crucial to the conceptual presentation of flames
—that they promote and guide both ideas and appropriate action.

The distinct advantage of doing a flame analysis of ECE policy is that it will
reveal the values and facts appealed to as policies were being developed and
implemented. In addition, because this is a retrospective examination of policies that
have in fact been enacted, it allows me as the researcher to select a set of policies that are
quite different and were enacted a different time periods to examine both differences over
time and potential differences in flaming connected to differences in these policies.
Methodology and Findings

Based on an initial study of the state of Michigan, I modified this research
program to make analysis of flames used in ECE more workable. For this study, I
examined State of the State Addresses flom governors in Oklahoma and Michigan flom
1980 to 2005. I looked at three things in these speeches: flames used for ECE, for
education and for welfare. I paid particular attention to the type of flames as categorized
by Snow and Benford (1986): diagnostic, prognostic and motivational. In order to situate
these flames in terms of spending in these three policy areas, I present spending data for
education and welfare and note where initial funding of ECE policy compares to these

trends.

111

These two states were selected because of the differences in their existing ECE
policies. Oklahoma is one of three states to provide universal preschool to all of its four
year olds putting it at one extreme of state involvement-full provision. Michigan is
somewhere in the middle of the pack of states funding a targeted preschool program, the
Michigan School Readiness Program (MSRP), as well as Project Smart Start - a modified
SMART START type collaboration and coordination program. Because these policies
were enacted at different times in the last 20 years, this set of state data also serves to
cover the range of years during which many federal and national changes were taking
place in the world of ECE. Figure 4.3 in the Appendix has a timeline reflecting this.
Oklahoma: A study of universal preschool enactment

The State of Oklahoma enacted statewide voluntary universal preschool in 1998
setting it apart flom all but Georgia (at the time) in provision and firnding of preschool
for all four-year olds whose families elected to place them in preschool (NIEER, 2007).
The act itself was put forth by an inexperienced legislator who was able to assemble a
quiet coalition of support for the legislation (Fuller, 2007). It was highly unusual that
Oklahoma, a relatively conservative state without a reputation for progressive policies,
would be a flontrunner in an area that had befuddled state leaders such as California and
New York. Nonetheless, it stuck — and now Oklahoma consistently has the highest
number of four year olds served of all the states — around 92 percent of the four-year olds
in Oklahoma attend publicly funded preschool (with 68 percent in the state program, 18
percent in Head Start and 6 percent in Special Education preschools) (NIEER, 2005).

Between 1980 and 2005, Oklahoma had five different governors. The political

party switched each time a new governor took office, establishing an interesting pattern.

112

Under Governor Nigh, a previously enacted pilot preschool program was run, but quickly
fizzled out. Under Governor Bellmon, the state made provisions that allowed for an
expansion of pre-existing Head Start programs. And finally, in 1998, under Republican
Governor Keating, Universal Preschool was signed into law. As Table 4.2 demonstrates
this happened when the state goverrunent was in fact divided — both political parties held
either a chamber majority or a govemorship.

Table 4.1: Oklahoma Governors, by Term, Political Party and ECE Policy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Name Term Political Party State Enactment of ECE
policy

Nigh 1980-1986 Democrat 1980; Pilot program —-
dropped

Belhnon 1987-1990 Republican 1990; Head Start
expansion

Walters 1991-1994 Democrat

Keating 1995-2002 Republican 1998; Universal Preschool

Henry 2003-2006 Democrat

Table 4.2: Oklahoma Governors by Term and Divided or Unified Government Status

Name Term Unified Divided

Nigh 1980-1986 "‘

Bellmon 1987-1990 *

Walters 1991-1994 *

Keating 1995-2002 *

Henry 2003-2004 *

2005-2006 *

 

 

 

 

 

 

Spending in Oklahoma on social programs and education has followed a distinct

pattern. From 1980 to 1990, education funding increased very incrementally rising flom

$1.6 billion to $2.4 billion (U .S. Census, 2004). After reaching a peak in 1998 at $3.9

billion, firnding hovered around $2 billion for most of the early 20008. This reflects

changes in state funding formulas primarily, demonstrating that the state “share” in

education in recent years has fallen. Welfare expenditures in Oklahoma grew

incrementally over this time period, with a more rapid increase in the last five years.

113

 

What is perhaps most interesting is that while spending in these areas has remained
relatively flat, the percentages of total expenditures of welfare and education have
dropped somewhat dramatically. State education expenditures were around 39 percent of
total direct state spending in 1980, but only around 17 percent in 2005. State welfare
expenditures were 13 percent of total in 1980 and jumped to above 20 percent between
2000 and 2005. State preschool funding falls squarely in this set of expenditures with

initial funding in 1999 of around $45 million.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Oklahoma: State Expenditures 1980-2005
16000
14000 —— .1
12000 —---—— ~ — —
g10000~ -- — ,_ A . ~m +We|fare I
g 0000 e——- -~ ~ . A total i1
5 6000 .m—w ———e ~ H - Var—Education!
.fl’
2000 F. ---..- «a t
0 1 ' I I l 1 H 1 l I 1 T I 1 T 1 1 l l 1
<9
of of of of of at" a” e‘t of e” vs” vs”

 

 

 

Figure 4.1: Oklahoma expenditures data for education, welfare and totals (US. Census
Bureau)

Governor Bellmon was the first Oklahoma governor to mention early childhood
education (ECE) in his State of the State address in 1987. It was covered in most, but not

all subsequent speeches including in seven of the next eighteen SOS speeches (1989,

114

1994, 2001, 2002, 2003, 2004 and 2005), though to varying degrees”. Table 4.3 lists the
major flames used in these speeches to present the issues of education, welfare and ECE.
Education and Welfare are certainly larger issue areas in size, scope and firnding,
however, once ECE was introduced it tended to be covered as either a welfare “child
care” issue or later as a school “issue” once universal preschool was in place. For the
most part flames used in education and welfare do not overlap — with exceptions being
the flames of “parents” and “investment”. However, while the “investment” flame in
education is usually motivational, in welfare it is more prognostic. ECE flames are more
limited partly because it is a program of much smaller scope and also likely because it is
a relatively new area of state activity. While flames for “child care” forms of ECE tend

to be “care” and “market”, those for preschool are focused on “achievement , research”

and “investment”.

Table 4.3: Oklahoma: Frames used in Education, Welfare and ECE

 

 

Education Welfare ECE

Frames Schools and Job Link Temporary Care
Funding Safety Research
Basics Social Costs of Poverty Private Market
School Choice/Market Family Values
Standards Reform Crime
Model for other States Responsibility
Efficiency Fraud
Accountability
Competition
Teacher
Achievement Achievement
Parent Involvement Parents Working Parents
Investment Investment Investment

 

 

 

 

 

 

Governors used flames to both establish links between programs and professed

outcomes (motivational) and to diagnose a problem (diagnostic flames) and propose its

 

‘4 Oklahoma State Archives. SOS speeches for Governors Keating, Walters and Nigh. Found on the
website at http://www.odl.state.ok.us/oar/govemors/home.htrn. SOS speeches for the current Governor
Brad Henry found the State of Oklahoma’s website http://www.governor.state.ok.us/index.php.

115

solution (prognostic flames). Interestingly, the flames in education were predominantly
motivational and prognostic, while the flames in welfare were more diagnostic and
prognostic. ECE flames tended to be motivational and prognostic. Some examples of
the verbiage used to do this “flaming activity” help demonstrate how these flames are
used and how each type “works”.

Table 4.4: Oklahoma Education Frames

1. Governor Bellmon (1989) “The highest priority for spending new funds is in the area of
education because, clearly, economic development is built on a solid educational foundation”. This is a
motivational frame in that it links education to economic development in hopes establishing support for
education funding increases.

2. Governor Keating (2002) “We must put I place a stronger remedial education program to help
those who lag behind, catch up and succeed”. This is a prognostic frame as it ties the problem (children
who are behind in school) to a solution (stronger remedial education programs).

Table 4.5: Oklahoma Welfare Fryames

1. Governor Bellrnon (1989). “$1.9 million in state dollars is being proposed to expand the
funding of Aid to Families with Dependent Children Program, to include two-parent families. It is
Shortsighted for the State to provide aid to individuals only after they have agreed to destroy their legal
family relationship. It is wise state policy to strengthen, not destroy, families. The current program ignores
the fact that families are now driven apart in hard times simply to receive survival aid.” This family flame
is prognostic, Bellmon asserts that the current program does not support families and proposes a solution -
funding to include two-parent families.

2. Governor Keating (2002). “If you can take care of yourself to be truly human, to participate in
this glorious event we call ‘life’, you have to take care of yourself. And as a result of the reduction of your
welfare roles (sic), we have many men and women who have never worked that are actively rrrembers of
the Oklahoma economy”. This is a motivational frame as it promotes the great success in welfare
programming in encouraging personal responsibility.

Table 4.6:gdahoma Early Childhood Frames

1. Governor Bellmon (1989). “All employers, public and private, must begin to accommodate the
changing character of the work force. Funding will be provided through the Commission on Children and
Youth to encourage the development of private sector day-care programs”. Again this is a prognostic
flame, establishing the “problem” as increased working families and providing a solution- funding for
private day cares.

2. Governor Henry (2004): “All Oklahoma families should have access to child care and early
education programs in their cormnunities. By expanding and strengthening early childhood programs, we
can achieve our state goal for the public schools of having 90% or more of third-grade students reading at

their grade level by 2007”. This “achievement” flame is also motivational, promising great results in
achievement later on with investments in early childhood now.

An examination of how Oklahoma Governors have flamed the issues of education,

welfare and ECE does seem to show evidence of different flaming activities for each

116

policy area with some overlap. Figure 4.4 in the Appendix shows a graphic presentation
of how these flames interact in these three policy areas.
Michigan: A Study in mixed state involvement

The State of Michigan enacted a targeted state preschool program beginning in
1989 to serve primarily children who were not eligible for Head Start, but were still flom
low-income families or had other risk factors deemed necessary for preschool
intervention. The MSRP program was piloted in 1985 and then under the leadership of
Governor Blanchard who fought for it in all subsequent years, it was enacted in 1988. In
2003, newly elected Governor Granholm promoted a “Project Great Start” legislation
modeled largely on North Carolina’s heralded SMART Start program that provides a
large amount of state funding to support collaboration and coordination of local ECE
efforts. Project Great Start was enacted in 2003, but it was not until 2005 when the
Governor formed the Early Childhood Investment Corporation to oversee and implement
it that it began to direct sizable grants to communities. Compared to Oklahoma,
Michigan’s MSRP program serves only a small amount of the state’s four-year olds (19
%) such that only 39 percent of the state’s four-year olds attend publicly funded
preschools (NIEER, 2005).

Between 1980 and 2005, Michigan had four different governors. As in
Oklahoma, the political party switched each time a new governor took office, establishing
an interesting pattern. Governor Blanchard was the primary champion for the MSRP
program and following him, Governor Engler did little to change the status quo for ECE
in Michigan, focusing on larger efforts at school funding reform, school choice and

welfare reform. Governor Granholm has championed ECE and added the weight of the

117

state’s resources to the ECIC in order to support ECE, but under extreme budget
shortfalls. Table 4.5 shows that both ECE programs in Michigan were enacted under
divided governments as both Blanchard and Granhohn worked with Republicans to
achieve these programs.

Table 4.7: Michigan Governors by Term, Political Party and ECE Policy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Name Term Political Party State Enactrnent
Milliken 1980-1982 Republican
Blanchard 1983-1990 Democrat 1985; Pilot MSRP
preschool; 1988; full
funding
Engler 1991-2002 Republican
Granholm 2003-2006 Democrat 2003; Project Great
Start
2005; ECIC
Table 4. 8: Michigan Governors by Unified or Divided Government Status
Name Term Unified Divided
Milliken 1980-1982 *
Blanchard 1983 *
1 984-1 990
Engler 1991-1993
1994-1996 *
1997-1998 *
1999-2002 *
Granholm 2003-2006 *

 

 

 

 

 

 

Spending in Michigan on social programs and education has followed an
interesting trend. Welfare spending has increased at a slow, but constant rate growing
flom $2.7 billion in 1980 to $9.4 billion in 2005. Unlike welfare firnding, education
expenditures have had a few dramatic changes over time. From 1980 to 1986 there was
steady growth (flom $6.4 billion to $9.4 billion). There was then a dramatic plunge in
state expenditures during the late 19808 and early 19908. Then after the passage of
Proposal A which changed the state school funding formula, there was a sharp increase

between 1994 and 1995 (flom $8.7 billion to $13 billion) reflecting the larger state share

118

in school funding. That was followed by a period of almost zero grth and then a
second “plunge” in 1999 due largely to the state’s economic downturn and cuts in
education expenditures. State preschool firnding falls squarely in this set of expenditures

with initial funding in 1989 of $25 million for MSRP.

 

 

 

 

 

Michigan: State Expenditures 1980-2005

 

 

 

 

 

60000
50000 'U—‘_#Aim * h - — —— ‘ ,9“ KT,
/X
.,, 40000 _ . -/‘———— ___
a Hat X + Education '
.2 / i
F 30000 4 A W _, — ‘ fix? 1 .. Welfare

 

E ’X/(\ as" .
M we? ’ ~+<-— Total

 

 

 

’\ 0., ’\ “a
0.1 0.1 Q §

 

 

 

Figure 4.2: Michigan Expenditures for education, welfare and totals (US. Census
Bureau)

Governor Blanchard was the first Michigan governor to mention early childhood
education (ECE) in his State of the State address in 1986. It was covered in most, but not
all subsequent speeches including in thirteen of the next nineteen SOS speeches (1987,
1988,1989, 1991, 1992, 1994, 1996, 1998, 1999, 2000, 2003, 2004 and 2005), though to

varying degrees15 . In two years, 1996 and 1998, Governor Engler did not address

 

15 Michigan Senate Journal. SOS speeches for Governors Milliken and Blanchard were pulled flom bound
copies of The Michigan Senate Journal held at the State Archives in Lansing, Michigan. SOS speeches for
Governors Engler and Granholm were pulled flom the State of Michigan’s website

119

preschool or “early childhood education”, but did address child care elements of his
welfare plan. Table 4.9 lists the major flames used in these speeches to present the issues
of education, welfare and ECE. Education and Welfare are certainly larger issue areas in
size, scope and funding, however, ECE shares several flames with these two issue areas.
For the most part frames used in education and welfare do not overlap — with exceptions
being the flames of “spending”, “parents” and “state role”. The “parents” flame in each
are is usually “prognostic” as the governors sought to profess what parents should be
doing either educationally for their children or in being financially responsible for them.
ECE flames overlap with those of welfare and education more in this Michigan sample,
than they did in Oklahoma. However, interestingly, the overlaps of ECE with welfare
and education are in areas that welfare and education overlapped already — possibly
pointing to a set of common flames in social policy more generally “spending”, “at-risk”,
“parents” and “state role” featured prominently in flaming all three issues. ECE did have

,9 6G

a set of exclusive flames that were not used in either other issue area: “readiness , early

99 6‘

learning , research/evidence” and “economic return”.

Table 4.9: Michigan: Frames used in Education, Welfare and ECE Policy

 

 

 

Education Welfare ECE

Frames Schools and Job Link Dependence Readiness
Accountability Opportunity Early Learning
School Failure Need/Helplessness Research
Parental Choice Responsibility Economic Return
Schools as Investments Model of Reform
Spending Spending Spending
At-Risk At-Risk At-Risk
Parents Parents Parents
State Role State Role State Role

Family Values Family Values

 

 

 

 

 

 

http://wwwmichigangov/gov/O,1607,7-168-23442_2198 1---,00.html, for Governor Granholm and
www.michigan.gov/documents for Governor Engler.

120

Governors used frames to both establish links between programs and professed
outcomes (motivational) and to diagnose a problem (diagnostic frames) and propose its
solution (prognostic flames). Interestingly, the flames in education were predominantly
motivational and prognostic, while the frames in welfare were more diagnostic and
prognostic. ECE frames tended to be motivational and prognostic. Some examples of
the verbiage used to do this “flaming activity” help demonstrate how these frames are
used and how each type “works”.

Table 4.10: Michigan Education Fraines

1. Governor Blanchard (1986): The “quality of our educational system determines the economic
strength of our state and our quality of life”. This is a prognostic flame proposing the solution of economic
weakness is quality education.

2. Governor Engler (1999): “the quality of our future depends on the quality of our schools. Better
schools and skilled graduates equal higher incomes and stronger families.” This links schooling to the
economy and to families and demonstrates a prognostic frame.

3. Governor Granholm (2004): “Never in history have businesses so badly wanted precisely what
we as parents want — highly-skilled , value-oriented citizens who will be successful in life and in the new
knowledge-based economy”. This is a motivational fi'ame, landing the alignment of business and private
family goals for education.

Table 4.11: Michigan Welfare Frames

1. Governor Blanchard (1986): “we cherish our responsibility to improve the lives of those with
legitimate needs. But as I said earlier, our strategy demands that ‘ government be innovative and efficient in
working to create independence, not dependence’”. This incorporates two flames a diagnostic “needs”
flame and a prognostic “create independence” frame.

2. Governor Engler (1999): “Another area where Michigan’s compassion has really made a
difference is our strategy to strengthen families. Thanks to our decisions and the efforts of our
communities, the number of people on welfare has fallen to the lowest level in 30 years.” This “family”
flame is prognostic, the problem was weak families, and the solution, welfare changes, has worked.

Table 4.12 Michigan Early Childhood Frames

1. Governor Blanchard (1986): ”There is a consensus nationwide, and we’ve seen it even in
Michigan, that early pre-school education may make the most differences of all in providing hope and
opportunity to our children. Head Start, I think proved that. That is one great program there is unanimity
on.” This contains both a motivational flame promoting the amazing benefits of preschool above all others.
It also contains the “evidence ”frame that sets ECE apart flom education and welfare.

2. Governor Engler (1999): “The launch of the READY (Read, Educate and Develop Youth)
program was successful. The goal this year is to make this reading readiness kit available statewide to
parents of young children. The goal of our strategy - for every child to be a good reader no later than the
end of third grade. . ..If you cannot read, you will not succeed”. This is another prognostic frame, linking
early reading with later success.

3. Governor Granholm (2004): “Last year we began a revolution in education when we publicly
declared that education in our state will begin at birth, not when a child enters kindergarten. Breakthroughs
in medical science have taught us that 85 percent of a child’s brain development occurs in the first three
years of life. It’s now an accepted fact and groups flom the American Academy of Pediatrics to the
Business Roundtable have recognized that we must act on this knowledge to give every child a great start

121

in life.” This selection contains both a motivational frame and demonstrates the “evidence ” or “research ”
flame that is unique to ECE in these speeches.

An examination of how Michigan Governors have flamed the issues of education,
welfare and ECE does seem to show evidence of different flaming activities for each
policy area with some'overlap. Figure 4.5 in the Appendix shows a graphic presentation
of how these flames interact in these three policy areas.

Analysis and Conclusiong

This study has produced modest evidence of flames used by governors in
education, welfare and comparatively in ECE. There are certainly common flames used
in both Oklahoma and in Michigan in each policy area. There is also some overlap
between them, though less in Oklahoma than in Michigan. ECE flames seem to be fairly
consistent at this point in time. There is a great deal of flaming around “investment”,
“readiness”, “family” and “parents”. One outstanding or unique flame that presents itself
in this analysis is that of “research” or “evidence”. This is a strong frame used to
promote ECE as a worthy policy and to demonstrate that it “works”. This flame appears
to be unique in this study to ECE.

The next step in this process of flaming would be to create a way to link flames to
policy outcomes. Furthermore, to link flames to implementation and child outcomes.
Those are projects for another time. A flame analysis such as this is limited in causality
and efforts to expand its use may prove fluitless. However, it does allow me to examine
how it is in the midst of all the differing information, goals and policies available, a

single policymaker — in this case a governor — selects flames to support ECE.

122

Chapter 5: A 50-State Diffusion and Innovation Study of ECE Policy

Using the state as a fundamental unit of analysis, political scientists have
examined the diffusion of policy innovation and have continued to improve the models
for further research. However, little is known about the effect of competing policy
domains on innovation diffusion. In other words, how does a single policy diffiise when
it arises and operates in different policy domains? This paper explores how a single
policy diffuses in the states based on the relative strength of the competing domains.

Policy innovation is defined in the literature as any policy that is “new” to the
state adopting it (Walker, 1969). Essentially, there are two theories of how state
innovation policies are adopted: internal determinants and diffusion. Internal
determinants models argue that it is the political, economic and social characteristics of a
state that determine when and if it will adopt a policy innovation. Diffusion models, on
the other hand point to factors external to the states that affect such adoption. These
models are the national interaction, regional diffusion, leader-laggard and vertical
influence models. Based on Berry and Berry’s initial work in this field using event
history analysis (EHA), recent researchers have been able to utilize both the regional
diffusion and internal determinants models in a unified way (Berry and Berry, 1990;
Mintrom 1997; Volden, 2004).

Yet, there has been less done to examine policy innovation in the context of
multiple policy domains. Glick and Hays argue that reinvention of policy and extent of
adoption (differentiation of deep and superficial adoption) are highly relevant to the
understanding of policy innovation, but are not reflected in current models. They

theorize that states reinvent innovations for many reasons, thus state laws differ in

123

important ways and that this impacts the breadth of research on diffiision of policy (1991,
p. 838). They do find that policy reinvention occurs along a “common dimension”, but
that particular provisions lead to reinvention in more than one direction (1991, p. 847).
This firrthers the understanding of innovation diffusion, yet their work focuses on a single
dimension when some policies may diffiise over two.

Given the complex history of ECE policy and the nature of the present expansion
into so many different policies and stakeholders, both diffusion and reinvention are
critical to understanding what leads states (now the dominant activity centers of ECE
policy) to take action. A diffiision model allows me to test for what factors within a state
effect adoption of policy and what factors external to it affect likelihood of adoption.
This chapter examines ECE policy diffusion in the American States. Through an event
history analysis of state adoptions of state preschool, it demonstrates the effect of two
state institutional domains - welfare and education - on a single policy. The evidence
supports the hypothesis that the relative strength of one policy domain -—in this case
education- directly impacts the adoption of a policy. In addition, I present evidence of a
great deal of reinvention by states (early adopters are more likely to reinvent) which has
accelerated in recent years. Finally, I present a brief study of non-adopting states and
what ECE looks for them-they have significantly greater participation in federal

programs.

State Preschool as a Policy Innovation

 

Policy innovation is defined in the literature as any policy that is “new” to the
state adopting it (Walker, 1969). To understand the impact of state institutions on policy

innovation, I will examine the adoption of state preschool programs. I will use the

124

innovation, I will examine the adoption of state preschool programs. I will use the
approach developed by Berry and Berry, but will add to it measures of institutional
strength. I do this by looking at the institutional choice of a state for state preschool
policy. The goal is to determine if competing institutions add a dimension to the diffusion
of innovation.

In the United States there are separate systems of education for young and old
children. Almost universally, children ages five to eighteen attend “school”, while
children younger than five have less predictable situations: childcare, home or
“preschool”16. At the state level these separate systems are reflected in the different
institutions and agencies that operate and regulate schools and those that operate and
regulate childcares and preschools. This is not to say that there is no overlap, interaction
or cooperation between the school and childcare systems. Certainly both serve children
and. their parents.

Recent K-12 school reform efforts, including standards based education and
federal accountability, have included preschool programming and fimding as an element
of K-12 school reform. This aligns with the steady push in the early childhood education
community for universal preschool because of the well-evidenced notion that preschool is
important to success later on and because universal preschool would legitimize childcare

as both an institution and as a profession. Both reflect the societal demand for quality

 

“5 For the sake of clarity it is necessary to lay out some of the terminology involved in this paper.
First of all, generally speaking “child care” refers to any non-parental care arrangement for any child
between the ages of zero and five. (For the sake of this paper, “child care” will not include the care of
school age children when not in school). Similarly, “early childhood education (ECE)” includes
programming for children ages zero to five. “Preschool” refers only to a center-based child care
arrangement for children ages three and four (sometimes S-year-olds are included). Children ages zero to
one are “infants”, one and two-year-olds are “toddlers” and three- and four-year-olds are “preschoolers”.
Finally, “universal preschool” means voluntary publicly funded preschool for 3 and 4 year olds.
(Schweinhart and Fulcher-Dawson, 2006).

125

The private market and informal arrangements have increased to meet some of the
demand for childcare and preschool, and the federal government has also been a long-
term actor in it, through the Head Start program.

State preschool programs as a policy innovation involve two separate institutions:
schools and childcare systems (largely through welfare agencies). When a state considers
funding a “state” preschool program it does so in the context of K-12 school systems and
their standards as well as in the context of the early childhood education community in
their state. One is a formal system, one informal, but they are intrinsically connected to
the other. In addition, because there are policies and needs at other ages besides
“preschool” age, policies aimed at preschoolers may also effect or include younger
children. In light of these institutional constraints, why and when do states adopt state
preschool programs? What types of reinvention happen once a policy is enacted? How
are non-adopting states using federal programs?

Data and Methodology

History and existing policy certainly affect the continuing development of
preschool. Historically, public aid for preschool and childcare has come primarily in the
form of needs-based programs Head Start and the Child Care Dependent Fund (or AFDC
or CCDBG its precursors). In contrast, most public aid for schooling -flom state and
local sources - has applied to and demanded equal treatment of all children regardless of
race, income or ability.

In order to test the effect of relative policy domain strength on adoption of state
policy, this current analysis uses a method, as introduced by Berry and Berry into state

policy and politics research, called event history analysis (EHA). This method is also

126

policy and politics research, called event history analysis (EHA). This method is also
deemed hazard analysis and provides a quantitative way to answer the question: Given
that a state has not adopted state preschool policy, what is the probability that it will do so
in a given year?

To proceed with this analysis, I collected annual data from forty-six17 states flom
1978 to 2000. The analysis starts in 1978 because that is the year Florida passed a state
preschool policy. I chose to start in 197 8 with Florida’s Migrant Preschool Education
law because it represents the first effort of a state to enact preschool policy after Head
Start had already been well-established. While California and New York had enacted
state policies in 1965 and 1966, these policies do not represent an innovation in the same
way that later ones do. I would argue that in the initial years following Head Start
passage, states simply supported Head Start, but did not fund preschool as state programs.
However, after Head Start had been around for over a decade a series of states began to
adopt state preschool policies.

The data involve discrete time periods18 (yearly data) rather than continuous time
intervals. My goal is to understand the variables leading a state to adopt a preschool
policy. I define my event of interest as passage of state preschool policy (it should be
noted that in a few cases, the passage of such policy was not immediately funded or
implemented, but lagged one or two years) (ECS, 2004). The dependent variable is state

preschool program coded as O in the years prior to adoption, 1 in the year a state adopts it

 

‘7 States not included are California and New York as they passed legislation simultaneous to Head Start
legislation, Nebraska as it has a nonpartisan legislature and Pennsylvania as it enacted a law that allowed
four year olds to be a part of the K-12 system in 1966, but this is not considered preschool programming,
rather is implemented as developmental kindergarten which is not relevant to this current endeavor.

18 Typically, a lifetime T is left or right censored if instead observing T we observe a finite nonnegative
random variable Y and a discrete random variable with values 0, l or 2. By definition, when A=O, Y=T,
when A=1, YT.

127

in the data set for each year represent the “risk set” for that year. The hazard rate for a
given year is the number of states adopting state preschool programs divided by the risk
set. The risk set is the set of all states that have not previously adopted state preschool
policy.

I use a Cox Proportional Hazard model, which works similar to a conditional logit
model with fixed effects. The model is used to estimate the probability p that a state (i)
will adopt a policy in a given year (i). The goal is to determine what factors affect the
hazard rate during a certain time period (See Allison, 1982 and Berry and Berry, 1990,
for further explanation of using EHA for diffusion analysis). I regress the political,
school, socioeconomic and diffusion variables against the hazard rate to establish a
baseline (See Mintrom, 1997). Following this, I use this model to measure the impact of
relative policy domain strength on the hazard rate.

The dependent variable in this model is a simple dichotomous representation of
state preschool policy adoption. I gathered this data flom two aggregate sources, the
Education Commission of the States and the NIEER Preschool Yearbook. Whenever
there was disagreement between the two or need for clarity, I utilized the websites of
each state to verify the dates of initial legislation. I define the parameters “state preschool
policy” to include state enactment of a program targeted specifically at preschool
populations. I do not include programs that are simply state supplements of Head Start
(although I do consider these if the supplement is actually not just a fLmding stream, but
also a state effort to support additional preschool programming). I do include policies
that were passed as either stand-alone legislation or as a part of a more comprehensive

bill either in education or childcare/early childhood education. In addition, I consider

128

bill either in education or childcare/early childhood education. In addition, I consider
only the year of first adoption of any state preschool policy without regard for size or
scope or for whether or not the state subsequently changed or reinvented the policy.19 (I
present reinvention by states in the next section).
Policy Domain Strength Variable

A primary measure of interest in this analysis is relative policy domain strength. I
am most interested in examining the impact of the relative strength of education policy to
welfare policy in the states. In order to measure this, I have calculated a simple ratio of
direct education spending to direct welfare spending by states over the years of interest.
The calculation is done so that the higher the number, the higher the fiscal effort of states
on education is in relation to welfare effort. It is clear that this ratio masks a host of
variations that may occur within states in these two policy domains, however, education
and welfare remain the two largest areas of state spending and as such, this ratio serves as
a proxy of the relative strength of their policies.

Policy Domain Ho: The higher the relative education to welfare spending in a
state, the more likely that state is to adopt preschool policy in a given year.

Diffusion Tests
A regional influence independent variable is included on the assumption that
states are more likely to adopt a state preschool program if neighboring states have done

so. This variable is simply a proportion of the states bordering a state that has passed

 

'9 In NIERR’s Annual Preschool Yearbook, they define ‘state preschool policy’ slightly differently than I
do here. As such, while they include Pennsylvania, they do not include Rhode Island or Florida under their
definitions. However, a primary purpose of their state preschool yearbook is to provide rankings based on
access and quality and it is under those restrictions that their list of states with preschool policies is slightly
different than mine. I do not focus on either access or quality for this analysis, rather focusing on state
preschool as an innovation based on adoption of any preschool specific policy.

129

diffusion and innovation literature finds this variable to be highly statistically significant
(See for example Mooney, 2001 for a detailed examination of this variable).
Explanations for such findings range from the idea that states learn flom their neighbors
to the argument that states compete with each other to implement policies.

Regional Diffusion Ho: The more neighbors a state has that adopt the policy, the
more likely a state is to adopt

Political Variables

There are several ways to measure the impact of politics on state policy diffusion.
However, preschool policy presents an interesting case, as it does not necessarily have
one or the other party as its champion. While Democrats may be seen to be more
supportive of social programs like welfare and education and Republicans more likely to
support limited government, preschool policy has moved flom the federal government to
the states. As such, policy is implemented in the states as a devolved policy that
promotes the notion of states rights also favored by Republicans generally. The
continued bipartisan support of Head Start at the federal level does much to support the
notion that preschool is either a bipartisan or a nonpartisan policy issue. In other words,
there are no strong assumptions about which party is more likely to enact state preschool
policy. As such, I employ a set of political covariates that will help demonstrate if there
is an ideological bias to this policy or if it happens regardless of ideology of citizens or
state governments.

I include a dichotomous control variable for unified versus divided government.
Here I code a government as unified if the govemorship, upper and lower houses of the
legislature are all held by the same majority party (1= unified). In contrast to other

measures, this measure does not differentiate control by party; rather it aims to test if

130

policy passes regardless of unification and is bipartisan (or nonpartisan). Practically
speaking a unified government always has a greater chance of passing any policy,
however, in this instance; unification may be sufficient, but not necessary for the passage
of preschool policy.

In order to test ideology and public opinion, I employ two measures developed by
Berry et al (2003). One is a measure of citizen ideology and the other of state
government ideology. Both involve using a continuous scale of ideology to calculate
annual “scores” for both citizen ideology (representing public opinion) and government
ideology. Both measures read flom right to left across the ideological spectrum, with
higher numbers representing more liberal ideologies.

Unified Government Ho: States with a unified government (both houses and
govemorship controlled by same party) are more likely to adopt (as it is easier).

Citizen Ideology Ho: The more liberal the citizens in a state, the more likely it is
to adopt any social policy—including E CE.

Government ideology Ho: The more liberal a state’s government, the more likely
it is to adopt any social policy-including E CE.

Social and Economic Variables

There are several different ways to capture the social and economic determinants
of a state adopting preschool policy. The effects of quality preschool seem to
significantly benefit children in their subsequent social, economic and educational well-
being. This is especially true for high-risk populations such as minorities and children
flom poor families (Barnett, 1996). I use the poverty rate for each state, though the data
for this variable was calculated by the Census in a different way for part of my data set.20

An additional economic variable I use is the labor participation rate of women which

 

2° Beginning in 1990, data collect from the Current Population Surveys using revised procedures and is not
comparable directly to other years. However, across states, data is comparable for each year in this dataset.

131

for this variable was calculated by the Census in a different way for part of my data set.20
An additional economic variable I use is the labor participation rate of women which
reflects the demand for preschool in general based on the need for care while mothers are
working outside the home.

Poverty Ho: The higher the poverty rate, the more likely it is to adopt ECE policy.

Working Women Ho: The higher the rate of working women, the more likely a
state is to support and adopt E CE policy.

School Variables

The educational variable I use is the per pupil expenditure in constant dollars.
This variable is one that demonstrates the weighted level of educational spending in each
state. While it does not reflect state proportion of such expenditures, the policy domain
variable explained above captures this in a different way. Per pupil expenditure is a
common variable to use in educational finance literature, though some argue it has little
direct relationship to achievement outcomes (see for example, Hanushek, 1991).

Ho: The higher the per pupil funding, the more likely a state is to adopt E CE
policy in order to “protect its investment

EHA Results

Table 5.1 in the Appendix displays the results of the Cox Hazard Regression for
the data on state preschool policy. The first model demonstrates the significance and sign
of the various covariates described above. Regional effects were negative demonstrating
that states whose neighbors had preschool policy were less likely to adopt preschool
policy. The political variables turned out as expected with unified government increasing

the log likelihood of adoption, though not at a significant level. The ideology measures

 

20 Beginning in 1990, data collect flom the Current Population Surveys using revised procedures and is not
comparable directly to other years. However, across states, data is comparable for each year in this dataset.

132

percentage of females in the workforce, the lower the likelihood of state adoption of
preschool policy. This variable might be made more accurate in future models perhaps
by using a variable for females in the workforce with young children. As expected, the
poverty rate variable was directly related to adoption and at a significant level. The state
per pupil spending amount demonstrates again a negative coefficient where the higher a
state’s per pupil spending, the lower the log likelihood of preschool adoption. This
baseline or null model serves primarily as a comparison model for Model 2 which tests
the “policy domain variable” of this study which sets it apart flom other diffusion EHA
tests in the literature. However, this model could be more fully specified by including
interest group variables, competing agenda variables (what other social policy variables
were on the docket the year these ECE policies were decided on) or possibly other
welfare benefits variables.

Model 2 includes the policy domain strength variable, which is found to be
positive and significant at the .01 level. Thus, the higher a state’s educational spending
relative to its direct welfare spending, the more likely it is to adopt preschool policy, all
else being held constant. The remaining control variables keep both their signs and size
of coefficients flom model 1 to model 2. Table 5.2 in the Appendix presents the hazard
ratios for model 2 only and shows the interpretation of such ratios similar to odds-ratios.
For the purpose of this current analysis is it clear that the relative policy domain strength
of education to welfare has a great impact on a state’s adoption of preschool policy.
What is less clear is how other state variables affect the adoption of ECE policy. Future
EHA models based on this project might include professionalization of the legislature,

percentage of female legislators, national actions related to ECE policy and percentage of

133

child care spaces available pre-policy adoption. As such, this set of EHA results should
be considered preliminary.
Reinvention of Preschool and ECE Policy in the Mes

A second way to capture the institutional impact on diffusion is to use a multi-
adoption model. This model would allow me to capture “repeat buyers” of state
preschool policy (Mahajan and Peterson, 1985, p. 48). Essentially, this would allow me
to capture in each year, both new adopters and repeat adopters. As a policy evolves over
time, it encounters reinvention and diffusion models have a hard time capturing this
(Glick and Hays, 1991). The multi-adoption model might allow for the incorporation of
the institutional variable in a different way. I would hypothesize that as the issue of
public preschool has evolved since the advent of Head Start in 1965, early adopters have
modified and reinvented their original policies to be more like the ones new adopters
enact. If these early adopters also change the institution that implements the policy, this
would be evidence of policy reinvention that Glick and Hays speak of, but also of an
institutional dimension to diffusion. For example, say state A is an early adopter of a
state preschool modeled on the needs based program of Head Start for four year olds and
the state family services welfare agency runs it. When states B and C adopt the policy,
but include three-year olds, state A may readopt state preschool, by simply adding three-
year olds to their existing program a form of simple reinvention. Altemately, if states B
and C adopt a universal preschool policy as a part of sweeping standards based school
reform, it would impact the reinvention of and the institutional choice of state A for re-

adoption.

134

Using the descriptive qualitative presentations of preschool and early childhood
programs provided by ECS and NIEER (as used above in the EHA calculations), it is
possible to discern both the level and type of reinventions happening in state ECE policy.
For this analysis, I have reselected all states (not dropping CA, PA, NY or NE which
were dropped for model specification in the Cox-Proportional model). What I present
below are tables and analysis of ECE policy reinvention described as adoption of
additional legislation once a state has previously adopted initial legislation in ECE. First,
I present level of reinvention which is simply a calculation of how many times since a
state adopted its initial ECE policy, that is has adopted new policy. This also involves
comparisons of early and late adopters and the flequency of policy adoption. Second, I
present the type of reinventions that states are adopting which allows me to differentiate,
for example, if a state is adding teacher quality standards, increasing the population
served and/or making funding formula changes.

Table 5.3: ECE Policy Reinvention in the States: 1965 toiresent

 

 

 

 

Total Reinventions By Average Number of
Adopters Reinventions
70 1.75

Early Adopters (pre- 42 2.0

1990) N=21

Recent 28 1.47

Adopters(post- 1 990)

N=19

 

 

 

 

 

While a total of 70 reinventions have been adopted by states that initially adopted state
ECE policy, early adopters were more frequent in reinventing ECE policy than more
recent adopters. This is likely due to the fact that as policy in this area evolved, early
adopters modified their programs to accommodate modern needs and political goals. In

addition, states reinvented policy along several dimensions of this policy area. The most

135

common type of reinvention was involving the number and/or type of children served,
seconded by fimding either a new funding stream or a change in funding formula.
Learning standards and coordination of programs were also prominent forms of
reinvention for these states. There is less obvious variation between early adopters and
recent adopters in these specific types of policy. In other words, while early adopters are
more likely to reinvent policy, they do so in all different aspects of the policy domain
along with their recent adoption peers.

Table 5.4: Types of Reinvention by States

 

 

 

 

 

 

 

 

 

 

 

 

 

“U 3" o ‘H

t g g ~i—7; 3

o , a. o

V’ 0 g on w .2 ‘a‘ m

8 is '6 G E a 8 3 .5 a

—B —= 0,3, "‘ 'c :5 One '9,

2 § 8 §§ 5 E5 80

U [— < ..J m LL. 0 m 0 ‘6.
States 31 10 10 21 24 6 16
Early 16 4 7 11 15 2 10
Adopters
Recent 15 6 3 10 19 4 6
Adopters

 

_S_tates without Preschool Policy

One final way to consider this policy area is to look at the states with no state
preschool policy. Given the significance of the regional diffusion variable in the EHA
model above, it is not surprising that a majority of such states are concentrated in one
area: the mountain-plains (Idaho, Montana, North Dakota, South Dakota and Wyoming
all fall on this list). But other states are perhaps more surprising, ranging flom New
Hampshire to Mississippi. In order to consider ECE policy in these states, I looked at
two sets of information: rates of usage of federal ECE programs and comparisons
between these averages and the US average. This information is helpful, because under

diffusion and innovation theory the role of federal policy on state adoption can either be a

136

pressure-valve where federal policy decreases a state’s perceived need for additional

policy or a snow-ball effect where federal policy creates momentum for a state to adopt

additional or complementary policy (See Shipan and Volden, 2004).

Table 5.5: Head Start and Special Education Percentage Enrolhnents by Non-Adopting

 

 

 

 

 

 

 

 

 

 

 

 

 

 

States
Head Start Head Start Special Special State Head
enrollment 4s enrollment 3s ‘ Education Education Start
Enrollment 43 Enrollment Supplement
33
Alaska 16% l 1 8% 4 Yes
Idaho 12 4 7 4 Yes
Indiana 9 5 7 5 No
Mississippi 38 25 6 3 No
Montana 23 15 6 3 No
New 6 3 7 5 No*
Hanrpshire
North 24 16 8 4 No
Dakota
South 20 15 9 6 No
Dakota
Utah 9 3 6 4 No
Wyoming 18 10 16 1 1 No
Average 18 10.7 7.2 4.9
US Avg 11.3 7.3 6.2 3.9
Z score 065* 0.47* 0.26”“ 0.43“ *significant
difference

 

 

 

 

 

 

 

 

As Table 5.5 demonstrates, the usage of federal preschool programs in states that
have not adopted their own preschool policy is varied. Most striking are the high
numbers of three and four-year-old children using Head Start in some of these states:
Mississippi, Montana, North Dakota, South Dakota and Wyoming. Mississippi stands
out in particular because of the 38 percent of four-year-olds reported to participate in
Head Start. Wyoming stands out as well as not only is Head Start enrollment well above
average, special education enrollment is as well for both four and three year olds. A
simple Z-score calculation of the means of this group of states in terms of federal

program usage and the US means in the same programs shows that there are significant

137

differences in usage in states not-adopting policy, however because of the outliers and

small sample size, this is due primarily to a few states, not all of the non-adopters.

Conclusions

 

The primary purpose of this chapter paper was to determine what if any effect
competing policy domains and their relative strength has on policy diffusion across the
states. In the case of preschool policy, it is clear that relative policy domain strength, as
measured by a ratio of direct state spending, has significant positive effect on policy
adoption. It remains to be seen and examined whether or not this holds true for other
policy areas where two or more domains compete for a policy. In addition, it is possible
that there are better measures to capture this concept than the ratio of direct spending.
However, the results here are promising in contributing to the further understanding of
policy diffusion and innovation in the states.

An additional conclusion flom this analysis is that, in fact, preschool policy does
appear to have little or no ideological ties. While the citizen and government ideology
measures both had positive coefficients, they translated into rather small hazard ratios
(2.5% and 3.4% more likely to adopt). This is reflective of the policy scene at the
national level with the constancy of bipartisan support for Head Start preschool
programming. Finally, in terms of education policy, it seems clear that states that spend
more on education relative to welfare are more likely to adopt preschool policy. As such,
policymakers wishing to garner support and develop programming would be wise to
consider preschool policy flom and educational perspective, rather than just as a needs-

based program.

138

Additional examinations of reinvention of state policy demonstrate that early
adopters of preschool policy are most likely to reinvent (adopt revised or increased
legislation) than those who have adopted more recently. In addition, states reinvented in
many areas of ECE policy: children served, teacher quality, accreditation, learning
standards, funding, comprehensive serves and coordination of programs. While changes
to the number and/or type of children served and funding were the most common type of
reinvention, the range and flequency of all the types demonstrates further the scope of
ECE policy.

While this analysis is promising for understanding policy diffusion generally and
preschool policy specifically, it is clear that there is more work to be done in this area.
This analysis was limited by not allowing for inclusion of state’s subsequent policy
adoption and not differentiating between the types of policies being adopted (in terms of
size or scope). In addition, some more in-depth analysis of the variables affecting,
preschool policy is warranted, as some of the covariates were surprising in terms of their
size and sign. Finally, research is needed on the impact of preschool policies on
preschool usage as well as the interactions of several federal educational policies (the
Nation at Risk Report and the No Child Left Behind Act, to name a two) with preschool

policy programming in the states.

139

APPENDIX

140

 

 

 

 

 

80.300068 :33 0::
0: 8880\000080—08 30: 050—98— 353 00.. 00:00.33
303000 0:00:00 :33 08 00 033 :0 m: 80: :0 :03? 00:: 0:000:00 0000005000
806:3 NO: 008 :0 :23000 800 :00 05 M0 000 30:28 088 .33 0503:”: “35:0 08:25:05-3: mfiflaxm 00.32
3:080:05 303:0 808$ 080m
0800 088 E8 505 083:8 0: :00 :30 3020 030053 .8080: 050m
5 30:03.. 3:00 :0 0:802: 8.80.30: 0:38 300m 0303003002
30:08:? 0:802: 00 0:0: 0:0 0:93 0.8050! 00000—030
02? 0.88003 00:0 0:00: 0:0 38:80: .«0 8008002 00:l0w0 H02 30.528
=330:m 8:205:00 308 $800005: S0: 0: 800:3
000000 3:28: 02335 80303 :0 :0000:::00_ 8308038 :0 $2000: 830:: 0:088:00
gm :0 0m: 3:395 308 .«0 0:23:80 0>:S:0:_Y00_:0> 80:8 0000 2:083 000003:
0: :00» E 080: :0 w::00 00 08303 :05 30:38 0300:0002
3:3 .«0 000: 00338 000.030 803080, 0:00:38: 0:0 0:0> 00000830 8020,» 00:08 :0 088: 00m 00 300:0: E
005008 0:0 0030:? :00: 30% 0:0 0:00: 0:0 00 0.080: 8: 8000044. 0:508 m 003000 5303005 5:03 8:02.002
355:0 _00::0>
8:00:30 00:23: 80:00.: :0 033.
0830700000.: 0:03:00 080: 0>0E
8.5000,: :0E00xm 00003053005 ,3 08303 00:00:": 3:300 0:0 8300:?» 3:3 800:3 8:003:00

385800000 00500::

8000::0_Smnm08:0>0.::00 3:0:

0:0 006:3 .3038: 8800.8 m

0:: 04:050.”.—

 

08: :0 “8&0 0000.0 3:038:00
0:? 303—0 8:883:80 008-80%

00:00:02: 3020 00:00:”:

E0380: 8
0.30053 .0000 3:380: .bBwEE/w

383880.:—

 

3580>0E 308-

 

 

 

 

0238. 5:3 0:38-
00:08. .8088- :08600.
08.: :0>0 5:3 0:38- 80:80:00 8088- 82000 :0 =00-
3:08.005 3:3 80:02.2: 3:? 08:3 000030 00:00: 30: 8503:: 3.50808 3:3 mfiaixm .2005 :0 3.:
38.55 win—0...: 880:5 0—50.02

 

 

 

5:03 mom 0: :0:00:3< 0830.5 0:0 00:005. 3:03 003,—. («0 30:03:80 :N 0308

141

020 E282 :00 :8 000: .08 080-20 80 8000000 82080 “0.0 2:00:

 

 

88:00. £000 00 660.6 8038 0:0 30:00: 3803
800608

mm 80:3 030:0: :00 m :05 63003 8003
8:800:

000 806000 :03 050800000 00: 0:0 8.00:0m
8:0“ 0:0: 00 0:00 0:06 000: 880360: 80:03
.08 00002 002 3 000 22 80 8 088
@0000 30:8 :0: 88m 000:

020.5569 8030.:

800 0080820 808000 008.2
3000 080820 0 28.0

An 83 0002.0

A. 83 :08 :20:

380 800.0

88: 080008 .008 00000
$8080 00335800 032
.0800 3000500000

8000: €000 €088 0-0.0
880 02085 208:

0.0.003

 

 

3: a 08-000 0800 :8 08:

600000 8: 3880803
0:0 36:08 8:00:00 600.:

0000:0306 2.60 00: 03:

80:: :03 I 03 8603.6 0000:3003
8033

30 08:00 ”3633 60:96
b36008: 6603

0:08:36Q E03003

380 0035-326:
88: 080.0
88: :8 030 08.388
:8: 80. 00000000
0830838033 :0 :03N

82 00 xi 0:00 :20:

8.0.003

 

 

060 660800

000: 80508 33:03 I :33
8006000

0:03 30 000-:088030Q 60.5

60:80 808 8: 00:00:33 3003
803 8003 30 8000: 60830003
802 8800: 3:033

0t0$§§0§ 8030.:

A000:
-8 0 0308 0.0 00050.0
0 30800 5320030
8000 0 5800880
008 E80 088 000.:

006.00%

 

 

0:3

cmmn

 

8085. 680—00

’
V

 

A

 

 

 

00H E0002

:8 0.3:

:8 0000.2:

142

 

5:080 Foam—om 000% 0>0:

.08 3.8 .0088”: :00008. 30: 8:082:08 308 9 030:0: 0x08
38.0: 0: 2 Ewsofi 30.5308 003 :05 0:3: :0

.«o 888 :8 b.8808? :0 8:008 35 $08008 308 8503::
0:0 03? 3830 80:00 0 :0050: 88080850 $803 0 08080.6

 

808800: 05 0:0 8080b 05
.8 00080—0; 0 $8020.88 >020: 0 8:380: mo $008: 08 :03. :00

080a :03» 0 mo 0m0m: ..o 0080: ”$0900 05 :0 900»?
0808080:
08000 0:0 toga :080w 2 :080 8 008080: $888.38 8
0.00:0: 0:00:080 0:0 088.8% $803808: .8 0.88008
So 00:88: 80 0080—: 0: 2 0>0: >08 0020:, 30: 0
8:2. .88
8080mm 0:0 30:9 8:00 we 800000: 05 Am 0:0 080:0
.«o 3:338: 05 Av 808%080 80:0 00:38:07. A0 0803
no 802 053 0:00: :0 mo 30:82:00 220 0:30 0:030:
.«o 8:803:80 :o M80530: 05 00338 823,838:
\0.:0m $0303.68 3 3808008: 300:0 m020>
088 :0 0:0 w::0>0_0 0:0 .wEmfiEfl: 00238 :SBumbNQEQ
050; “0.00:0: :o m0:_0> 20: £800.80 8 8500.50
:080 :3 0:03 :0 mo 080d :0>_w 0 mflfiwdfiwfi 80 w::0>0_0 o
000: 820500 0:
88580 .3 00:0: 8 8080208 05 55 00006800 b05230
8: 0000008 805080 00098080 3 m0 8.. 808268
0 90 0083.00 0:0 20% 05 00830—0 0:0 0:080 80002 c
8800
.3 :03 0:0 8:30:00 30: 0 55 0:0: :0 .3 80:30:09
9.080368 8:00: 2 030 8 088 0 :0 88030808 0
.8088: 08 :0 008800 05

 

0080.:

8:05:00 Bo: 002:0: 86:00:00 08 :03? 8 8:808 8800:

.3 008008 05 03 88280: 8 0:03:00: x8258.“ 0:0 :05 085
A00:0E>0 “0:0 :00.“ 8 £0083 00:58:. 0080000: .3 00800. 0: :00 005

 

8:80:

830 38:30
A839
0830 0:088Q 8:80.: 9 @6004 8383.8an .0350.—
mmoooa E08088.“
05 30.088 :o 088me
00:08:03:
5:088:80; o
3
M
:0300E_q8< o
88:08”: 0
€8.80m 0:0 30:9
wawwtm o ...:o 080:0: 880d .3 :30003... 2E.
0:08:92 880003000002.— "880.:—
m08m06 5:0:
0:0 0:08 BE: 0080 888.: :8 :00: 0:9
30% 008028800 5:0: A33 80% 0:0 :wnomv
A09: 080802806 >020: 00.8.05 A055 .8 8&8
Ana—88...“; $80.... .3 8500803: 6:0 00000.5 2:.

 

38:80qu Roam—om EH0 00000005 $880.5 .m0:80>ob:oo how—om ”2.: 0BOH

 

.GOEUN ~8fl0§o>ow MO

38:09 :0: 3:30:05: 0:: :85 w:0_: £33000: 0:: ::0:: :8: 8:: 80:3:
0:: @802 0:_: 3:: $0888: 3:0: 88:00 2080:: :80 :0: 0:: :3:

 

 

:0 358085 50:00: 8 ::0w::: 8:: :09>:0: ::808:.:0>0w :08: 3:0:—9:1“: 0::00:m
03:: 0:: :0 :09:0:0: 0:: 80: :30: 38 3:: 33:88:
3:0: 3: :0 38080800 :0 :088 85:08: 0:: 0: 9:: :: 8:88.: 09:: 88m .3083
50:: 33:0: 880880>0w :08:::: 0:: :0 :0::00: :::: 8:808:08: ::: 80:00 08:: :0 80800 :0 00:»0: mm
8.5.3808
:0 _:0E:::mo0w ._:0::0:::::I 38800 080: 8 80:: 8:02: 398:: 08:: 880: 0:805. 63:580.»:
:00:::::: 0900:: :3 :0:::: :0 80:88:: 090:8: 0:»: 092030 :30: 8:8 mm
:08:::

00:08: 3:0: w::8:0:::

:0:0:-00.:0:::
3:0: :0 0:8085
:08:: 30:90::

$3: .50: :5 526::
02:08:: 3:0: :0 .8:—

 

 

 

0:::0_: 8 808:»: :8: 0:808: 0:: $8108: I 03:0: 3:0: 533-8800 w:::::: 0.6:? mm
.9: .: .32 .53: :5 55:?
:08:: :09: :8: 30:90:: 80: 0:::: 3:: :0::0:: 3:0: 0:: :0 :00: 0:: :: 0::
.::0::::08 0>:::0:0w :89: :0N8:w:0 .:08:: 3:882:02: .A:0::::08w::
.> ::0:0_0:B 88:9: .> 3.8:: ”080 .> 8:39 .w.0 .0:0:: :0 80:80:98
_:00_ 0:: :08:: 8:0: 3:09:98: :0:0: :0 88:3 :0:::: 3:880 93:: 3:830:82
:0:::: 3:0: 68800 0: 0:: 80:0: 3:09:98 :08:: 80:0: :05: _::09:.:::: $3: .80: :8: 5:09
893:: 3:0: : :0 8030:: 0:: 88800 0: 0:: ::0:0: _::09::::: 3:0: 3:69:80 8:05:03: um
:32 .:0:8:Ov 88:: :0 80:0:800
3:0w: 0:: :2: £8 88:0 80:0: 8 w::w:w:0 :0: 0.80:8 : .:8:: 0: ::0 : _::0::>.:0E
80:: 832:0: :0: 8:08 :8: :_:0m 093:: :8: 4.030:
0: 38 8030:: : 30: :0 8.5033 : 09:08 3:: 3:: 8 a380:0 89:0: 0::0:mo:: 6mg fi:0::0m :8: 30:5
8030:: 0:: :0: 08:3 :33: :8: 8:: :: 8030:: 0:: 08:0: 50: 0::0:w::Q 08:: :0 80:8: .5:
:0:&:0:0D 08: Z :0::0_:_:::_0 088m

 

80:89:35 :8: ::09::0Q m::8::m 0:::: ”3 .v 0_::H

144

1.1913 19919 mafmd [w

2000

[OOLpSQJd [BSJQMUQ )[0

1995

l 990

"“3 d‘dSW [W

lolld (MSW 1W g

1980

11918 "“3 ‘H'IDN

sneanzld pun %09 saqoeax uaxpqu
SunoA (mm uamom Buppom

1111039)] amglam

11918 Dean £1193

V'IWJ

[ooqosaxd
u; splo nae/C 17 go %09 amp 910w

08033

)I9Jd VElCII

38:11 12 new v

l°0q0591d 1 am

145

Figure 4.3: Timeline of 2-state data and Federal policies: 1980-2001

mom was 03:03 483325 8m magi «80520 #6 onE

 

 

 

 

Ea: moZmE>mEom<mmmm
82: Ave 8% as

  

82220 ham

 

 

 

 

 

 

 

 

 

 

 

 

 

/
=
mos—«.5
25m
EoEo>oEo< 88:0
BEBE“ I
055 \\|]: /. 52288
. AI". 88m
0.5.303 “comm?” \ fiomwmofifim—
mean:—
.858:
Uflwunm “V kum~m£3§000< / mDOH fig mommwm

 

 

£80 32%
/|Il\ floonom

852595

146

stowed
we .08:

 

823/
3E5

 

 

 

03:03

oocovaoaon

mom 28 03:03 .cozmosnm Sm 889m sawEomE ”3V 2sz

 

 

mUmeu>mEUm<mmmm

 

 

//
A08 858:8 sesame

823:6 ram

.7

 

 

 

 

 

Alli 20M 38m

5353550

3553
VERA/w
wficfimm
8.52,.—
any.

 

 

 

 

/
233
05 E
Eogmgfi 86:0
35:3
/ Showed
_oo:om
\ cosmoswm
£33582 / 30.35
Econom

147

 

.x; E Emu—Mammy: exam 8 Eda—mama: xxx: 3 “macadamia. .momofinoSQ 5 2m $8.5 Envy—Sm

 

 

 

fl fimom Nod g m Beam—8:4 mod
woo woo 3233030
6%.: 28.:
Nmmd- Nu _ .o- @659. Eng Mom 5:335
888 68.8
.3386 .886 >233 .x.
33.8 8.8.8
32:6- .12 fie- 88.38? E 838$ o\o ogoaeuoemuem
85.8 c 58
11.986 ...Immod $282 EoEEoBo
$5.8 38.8
256 omod xwofiog aoNEU
€38 €38
Need Need BaED .35....—
888 A33
mood- o 56- acute Ecommom Sam—£5
888
1:5de ......... 03:03 “cosmos—um 59.2% 555.5 hum—em «.553—
§ 3

 

8885 5:8 3285 we Sumo? 8% ”2 2.3.

148

..8 “3808881... 5% a “888%: $2 8 “8088.8...

 

 

 

:wom cooszoxfi w3

80 8283890

ovfio maximum 2&5 .Bm 253:3—
388 22: 8. 3H :83 €88 x

box: mmB $Nd— immwd 883.83 E 838$ Ax. oieaeuuemoew
388 use 8%.». 3.58; .8282 868080
mg; $282 SNEU

3m . _ BuED Roam—em

Sad 88% Reamom SEES

be; 29: 8% 5.88; 2833 8888 59.25 528: 88.— 88.3—

8888 8 82:85 N .982

 

deeming

.Amoumm Emwmmv xozom .8585 .«o gouge/x 88m ”Nm 2an

149

REFERENCES

150

Allison, P. D. 1984. Event History Analysis: Regression for Longitudinal Event Data.
Newbury Park: Sage.

Barnett, W. S. 1996. Lives in the balance: Age-27 benefit-cost analysis of the High/Scope

Perry Preschool Program (Monographs of the Hi gh/Scope Educational Research
Foundation, 11). Ypsilanti, MI: High/Scope Press.

Barnett, W. S. and LN. Masse. 2003. Defining program quality. In Debby Cryer &
Richard M. Clifford (Eds). Early childhood education and care in the USA (pp.
137-166). Baltimore, MD: Paul H. Brookes

Barnett, W.S., J. T. Hustedt, K. B. Robin, and K. L. Schulman. 2006. The state of
preschool: 2006 state preschool yearbook. New Brunswick, NJ: National Institute
for Early Education Research, Rutgers University.

Barnett, W.S. and LN. Masse. 2002. Funding Issues for early childhood care and
education programs. In D. Cryer and R. Clifford (Eds), Early childhood
education and care in the USA. Baltimore, MD. Paul H. Brookes Publishing Co.

Benford, R. and D. Snow. 2000. Framing Processes and Social Movements: An
Overview and Assessment. Annual Review of Sociology 26: 611-639.

Berry, F .S. and W.D. Berry. 1999. Innovation and Diffusion Models in Policy Research.
Chapter 7 in the book, Theories of the Policy Process, Paul Sabatier editor.
Westview Press: Colorado.

Berry, ES. and W.D.Berry.1990. State Lottery Adoptions as Policy Innovations: An
Event History Analysis. American Political Science Review 84: 395-415.

Besharov, D. and C. Hi gney. 2006. Federal and State Child Care Expenditures (1997-
2004). Maryland School of Public Affairs, Welfare Reform Academy. College
Park, MD.

Bureau of Labor Statistics 2005. Women in the Labor Force: A Data Book. US. Bureau
of Labor Statistics. Washington, DC.

Butler and Gish. 2006. The Child Care and Development Block Grant: Background and

Funding. Congressional Research Service. Library of Congress. Washington,
DC.

Campbell, F.A., C. Ramey, E. Pungello, J. Sparling and S. Miller-Johnson. 2002. Early
childhood education. Young adult outcomes from the Abecedarian Project.
Applied Development Science 6: 42-57.

Child Care Bureau. 2006. CCDF Data Tables. US. Department of Health and Human
Services. Washington, DC.

151

Children’s Defense Fund. 2000. Quality Child Care is Key to School Readiness.
Children’s Defense Fund. Washington, DC.

Clarke, 8., and F. Campbell. 1998. Can intervention early prevent crime later? The
Abecedarian Project compared with other programs. Early Childhood Research
Quarterly 13: 319-343.

Coburn, C. 2006. Framing the Problem of Reading Instruction: Using Frame Analysis to
Uncover the Microprocesses of Policy Implementation. American Educational
Research Journal 43: 343-3 79.

Cohen, March and Olsen. 1972. A Garbage Can Model of Organizational Choice.
Administrative Science Quarterly. 17 (1).

Cohen and Sardell. 2003. Policymaking for Children’s Issues. In Policy Currents:
Newsletter for the Public Policy Section of the American Political Science
Association. APSA 12(1).

Cohen, D. and Spillane, J. 1993. Policy and practice: The relations between governance
and instruction. In S. Fuhrrnan, (Ed.), Designing coherent educational policy:
Improving the system (pp. 35-88). San Francisco: Jossey Bass.

Coleman, J. 1966. Equality of Educational Opportunity Study (EEOS), ICPSR06389-v3.
Washington, DC: US. Department of Health, Education, and Welfare

Davies, S. 1999. From Moral Duty to Cultural Rights: A Case Study of Political Framing
in Education. Sociology of Education 72: 1-21.

Education Commission of the States. 2005. Early Learning: Prekindergarten Database.
State Profiles retrieved Fall 2005 and Spring 2006 online at:

http://www.ecs.org/ecsmain.asp?page=/html/educationlssues/ECSStateNotes.asp

Education Week. 2007. Quality Counts 2007: From Cradle to Career. Early Learning
Section. P. 38 of hard copy.

Frant, H. 1991. Specifying a Model of State Policy Innovation. American Political
Science Review 85: 571-80.

Friedman, M. 1962. Capitalism and Freedom. Chicago: University of Chicago Press.

Fuller, B. 2007. Standardized Childhood: The Political and Cultural Struggle over Early
Education. Stanford University Press. Stanford, CA.

Gamson, W.A. 1992. Talking Politics. New York: Cambridge University Press.

152

Gamson, W. A., and A. Modigliani. 1989. Media Discourse and Public Opinion on
Nuclear Power: A Constructionist Approach. The American Journal of Sociology
95:1-37.

Glick, HR. and Scott P. Hays 1991. Innovation and Reinvention in State Policy Making:
Theory and Evaluation of Living Will Laws. Journal of Politics 53: 835-50.

Goffinan, E.1974. Frame analyses: An essay on the organization of experience. Boston:
Northeastern University Press.

Green, D. E. 1987. Shaping Political Consciousness. Ithaca: Cornell University Press.

Hanushek, E. 1986. The economics of schooling: Production and efficiency in public
schools. Journal of Economic Literature 24:1 141-1177.

Harrington, M. 1962. 1981ed. The Other America: Poverty in the United States. Collier
Books: New York.

High Scope Educational Research Foundation. 2004. Significant Benefits: The
High/Scope Perry Preschool Project. See
http://www.highscope.org/Research/PerryProject/perrymainhtm

Head Start Bureau. 2006. Head Start Program Fact Sheet. Administration for Children &
Families. US. Department of Health & Human Services. Online at:
www.acf.dhhs.gov/programs/hsb

Head Start Bureau. 2006. Head Start Program Fact Sheet, Fiscal Year 2006. Found
online at http://www.acf.hhs.gov/programs/hsb/research/Z006.htm

Heckman, J. 2006. The Productivity Argument for Investing in Young Children.
Working Paper 5, Invest in Kids Working Group. Committee for Economic

Development.

Iyengar, S. 1996. Framing Responsibility for Political Issues. Annals of the American
Academy of Political and Social Science 546: 59-70.

J acoby, W. 2000. Issue Framing and Public Opinion on Government Spending.
American Journal of Political Science, 44, 750-767.

Kingdon, J. 1995. Agendas, Alternatives and Public Policies. Second Edition. Longman:
New York.

Kirp, D. 2007. The Sandbox Investment. Harvard University Press. Cambridge, MA.

153

Lieberman, R. C. and G. M. Shaw. 2000. Looking Inward, Looking Outward: The
Politics of State Welfare Innovation under Devolution. Political Research
Quarterly 53: 215-240.

Lipsky, M. 1980. Street-level bureaucracy: Dilemmas of the individual in public
services. New York: Russell Sage Foundation.

Mahajan, V. and RA. Peterson. 1985. Models for Innovation Difi’usion. Newbury Park:
Sage.

March, J. 1994. A Primer on Decision Making. New York: Free Press.

March, J. and J. Olsen. 1983. Organizing Political Life: What Administrative
Reorganization Tells Us About Government. American Political Science Review
77: 281-297.

McDonnell, L. 2004. Persuasion as a policy instrument. In Politics, persuasion, and
educational testing. Cambridge: Harvard University Press.

McDonnell, L. and R. Elmore. 1987. Getting the Job Done: Alternative Policy
Instruments. Educational Evaluation and Policy Analysis 1:133-152.

Millot, MD. 1994. Autonomy, accountability and the values of public education: A

current assessment of charter school state statutes leading to model legislation.
Santa Monica: RAND.

Mintrom, M. 1997. Policy Entrepreneurs and the Diffusion of Innovation. American
Journal of Political Science. 41: 738-70.

Montie, J ., Z. Xiang and L. Schweinhart. 2006. Preschool experience in 10 countries:
Cognitive and language performance at age 7. Early Childhood Research
Quarterly 21: 313-331

Mooney, C. Z. 2001. Modeling Regional Effects on State Policy Diffusion. Political
Research Quarterly 54: 103-124.

Moynihan, D. P. 1965. The Negro Family: The case for national action. US. Department
of Labor. Washington, DC

Mulligan, G., D. Brimhall, J. West and C. Chapman. 2005. Child care and early
education arrangements of infants, toddlers, and preschoolers: 2001. (NCES
2006-039). Washington, DC: National Center for Education Statistics, US.
Department of Education. Available online at nces.ed.gov/pub52006/2OO6O39Qdf.

National Access Network. 2006. Preschool Education: Overview. Retrieved from the
website at http://www.schoolfunding.info/policv/ece/ece.php3

154

National Association for the Education of Young Children. 2005. Critical Facts about the
Early Childhood Workforce. http://www.naevc.org/ece/critical/factSB.asp

National Center for Education Statistics. 2003. Prekindergarten in U. S. Public Schools:
2000-2001. US. Department of Education. Washington, DC.

National Center for Education Statistics. 2005. NCES Statistics in Brief: Child Care and
Early Education Program Participation of Infants, Toddlers, and Preschoolers.
October 2001. NCES See http://nces.ed.gov/pubs95/web/95824.asp

National Center for Education Statistics. 2005. National Household Education Survey,
Parent and Family Involvement. US. Department of Education.

National Conference of State Legislatures. 2007. Early Childhood Education Position
Paper. Retrieved online April 2007.

National Governor’s Association. 2005. Final Report of the NGA Task Force on School
Readiness: Building the Foundation for Bright Futures. National Governor’s
Association. Washington, DC.

National Institute for Early Education Research (NIEER). 2005. 2003, 2004, 2005, 2006.
Preschool Yearbooks. See http://nieer.org/resources/facts/index.php.

National Institute for Early Education Research (NIEER). 2004. Fast Facts. See
http://nieer.org/resources/facts/index.php.

National Institute for Early Education Research. 2002. Poll Shows Voters Want States to
Fund Quality Preschool for All 3- and 4-Year Olds. NIEER. Found online at
http://nieer.org/mediacenter/index.php?PressID=6

National Institute of Child Health and Human Development 1998. The NICHD Study of
Early Child Care. US. Department of Health and Human Services, National
Institutes of Health (Pub. No. 98-4318). Washington, DC: National Institutes of
Health.

National Research Council and Institute of Medicine. 2000. From neurons to
neighborhoods: The Science of early childhood development. Committee on
Integrating the Science of Early Childhood Development. J .P. Shonkoff & D.
Phillips (Eds). Commission on Behavioral and Social Sciences and Education.
Washington, DC: National Academy Press.

Nelson, T, Oxley, Z. and Clawson, R. 1997. Toward a Psychology of Framing Effects.
Political Behavior, 19, 221-246.

155

Peterson, P. and M. Rom. 1989. American Federalism, Welfare Policy and Residential
Choices. American Political Science Review 83: 711-28.

Plank, D. N. 1987. Why school reform doesn’t change schools: political and

organizational perspectives. Politics of Education Association Yearbook I 98 7. .
143-1 52.

Plank, D., and W. Boyd. 1994. Antipolitics, Education and Institutional Choice: The
Flight from Democracy. American Education Research Journal, 31: 263-281.

Rebel], M. and Hunter, M. 2006. The Right to Preschool in Education Adequacy
Litigations. National Access Network. Teacher’s College. Found online at
http://www.schoolfunding.info/resource_center/issuebriefs/preschool.pdf

Reynolds, A. J. 2000. Success in Early Intervention: The Chicago Child-Parent Centers.
Lincoln, NE: University of Nebraska Press.

Reynolds, A.J., J. Temple, D. Robertson, and E. Mann. 2001. Long-tenn effects of an
early childhood intervention on educational achievement and juvenile arrest: A 15-

year follow-up of low-income children in public schools. Journal of the American
Medical Association 285: 2339-46.

Reynolds, A.J., J. Temple, D. Robertson, and E. Mann. 2002. Age 21 cost-benefit
analysis of the Title I Chicago child-parent centers. Educational Evaluation and
Policy Analysis. Vol. 4: 267-303.

Rochefort, D. and R. Cobb, Eds. 1994. The Politics of Problem Definition: Shaping the
Policy Agenda. Lawrence: University Press of Kansas.

Rolnick, A. and R. Grunwald. 2008. Early Childhood Development Yields High Returns.
FRB Minneapolis. Federal Reserve Bank Spring 2008.

Schon, DA. and Rein, M 1994. Frame Reflection: Toward the Resolution of Intractable
Policy Controversies. New York: Basic Books.

Schweinhart L. and Fulcher-Dawson, R. 2006. Investing in Michigan’s Future: Meeting
the Early Childhood Challenge. Education Policy Center: East Lansing,
Michigan.

Schweinhart, L., J. Montie, Z. Xiang, W.S. Barnett, C. Belfield, and M. Nores 2005.
Lifetime effects: The High/Scope Perry Preschool Study through age 40.
Ypsilanti, MI: High/Sc0pe Press.

Sedlak, M., C. Wheeler, D. Pullin and P. Cusick. 1986. Selling Students Short:

Classroom Bargains and Academic Reform in the American High School. New
York: Teachers College Press.

156

Shipan, C. R. and C. Volden. 2004. The Influence of Local Laws on State-Level Policy
Adoption. T 0 be published

Snow, D.A., Rochefort, E.B., Worden, S. and Benford, R.1986. Frame Alignment
Processes, Micromobilization and movement Participation. American
Sociological Review 51 : 464-481.

Soss, J ., S. Schram, T. Vartanisan and E. O’Brien. 2000. Setting the Terms of Relief:
Explaining State Policy Choices in the Devolution Revolution. American Journal
of Political Science 45: 378-395.

Tyack, D. and L. Cuban. 1995. T inkering Toward Utopia: A Century of Public School
Reform. Cambridge: Harvard University Press.

True, J ., Jones, B. & Baumgartner, F. 1999. Punctuated-Equilibrium Theory:
Explaining Stability and Change in American Policymaking. Chapter 5 in the
book, Theories of the Policy Process, Paul Sabatier editor. Westview Press:
Colorado

US. Administration for Children and Families. 2005. Head Start Impact Study: first year
findings. Washington, DC:
http://www.acf.hhs.gov/programs/(mre/hs/impact study/reports/firstjr finds/firs
t__yr_f1nds.pdf

US. Census Bureau. 2008. Statistical Abstract of the United States. Washington, DC.

U.S.Census Bureau. Survey of Income and Program Participation. 1996 Panel, Wave 10.
See http://www.census.gov/population/www/socdemo/child/ppl-168.html

US. Department of Education. National Center for Education Statistics 2003.
Prekindergarten in US. Public Schools: 2000-2001. NCES 2003-019.

United States. 1970. Comprehensive preschool education and child day-care act of 1969.
Hearings, Ninety-first Congress, first and second sessions, on HR. 13520.
Washington: US. Govt. Print. Off.

Vinovskis, M. 2005. The Birth of Head Start: Preschool Education Policies in the
Kennedy and Johnson Administrations. University of Chicago Press: Chicago.

Vinovskis, M. 1999. History and Educational Policymaking. Yale University Press:
New Haven.

Vinovskis, M. 1995. Education, Society and Economic Opportunity. Yale University
Press: New Haven.

157

Walker, J. L. 1969. The Diffusion of Innovations Among the American States. American
Political Science Review 63.

Welch, S. and K. Thompson 1980. The Impact of Federal Incentives on State Policy
Innovation. American Journal of Political Science 24: 715-29.

Wells, A.S., Ctrutzik, Camochan, Slayton and Vasudeva. 1999. Underlying Policy
Assumptions of Charter School Reform: The Multiple Meanings of a Movement.
Teachers College Record 100.

Woolley, J. and G. Peters. 2008. The American Presidency Project: Political Party
Platforms Database. Available online at
http://www.presidency.ucsb.edu/platforms.php

Woolley, J. and G. Peters 2008. The American Presidency Project: State of the Union
Database. Available online at http://www.presidencv.ucsb.edu/sou.php

Wong, Kenneth K. 1989. Fiscal Support for Education in American States: The ‘Parity-
to-Dominance’ View Examined. American Journal of Education 97: 329-357.

Zahariadis, N. 1999. Ambiguity, Time and Multiple Streams. Chapter 4 in the book,
Theories of the Policy Process, Paul Sabatier, editor. Westview Press: Colorado.

Zill, N., Resnick, G., Kim, K., O’Donnell, K., Sorongon, A., McKey, R. H., Pai-Samant,
S., Clark, C., O'Brien, R., and D'Elio, M. A. May 2003. Head Start FACES
(2000): A whole child perspective on program performance - Fourth progress
report. Prepared for the Administration for Children and Families, US.
Department of Health and Human Services (DHHS) under contract HHS-105-96-
1912, Head Start Quality Research Consortium's Performance Measures Center.

158

   

1111111111111111