GOVERNMENT 2306(II) ESSAY

Policy Feedback and Preschool Funding in the American Statespsj_359 217..234

Andrew Karch

The early childhood education policy community has been described as a “divided constituency” in which groups with the same underlying goals sometimes work at cross purposes. This article examines how this internal division affects the contemporary funding of preschool education. It finds that states with a relatively large Head Start community are significantly more likely not to fund preschool education and significantly less likely to dedicate preschool funding exclusively to a freestanding state program. These results suggest that the creation and political solidification of Head Start generated policy feedback. They contributed to an ongoing tension within the early education community as Head Start beneficiaries viewed the creation of a freestanding preschool program as a political threat. This political dynamic illustrates the more general way in which the existence of a public policy can alter the dynamics of future political action.

KEY WORDS: policy feedback, preschool education, Head Start, state politics, interest groups

Introduction

In recent years, developments in neuroscience research and the efforts of a strong and growing advocacy community have helped place preschool education on the political agenda. Supporters of publicly funded preschool have experienced considerable success at the state level. Most states currently dedicate public funds to preschool education, yet several states do not. In addition, states that operate publicly funded preschool programs employ a variety of policy approaches. This article at- tempts to explain why states take divergent approaches to the funding of preschool education by isolating the political determinants of this policy decision. It argues that understanding the contemporary politics of preschool education requires an aware- ness of the history that has shaped, and continues to shape, the present terrain of preferences and actors (Pierson & Skocpol, 2002, p. 701). The preschool advocacy community has been characterized as a “divided constituency” (Michel, 1999) in which groups with the same underlying goals sometimes work at cross purposes. State officials’ decisions about whether and how to fund preschool education reflect this internal division, with the size of the Head Start community seeming to play an influential role.

The Policy Studies Journal, Vol. 38, No. 2, 2010

217

0190-292X © 2010 Policy Studies Organization Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford, OX4 2DQ.

The Decentralization of Preschool Education in the United States

In focusing on preschool education, this article emphasizes a subset of programs that frequently fall under the broader headings of childcare or early childhood care and education. Preschools are “programs offered under public and private education auspices or providing compensatory education under special legislation, usually serve children from age 2–3 to compulsory school entry, are largely half-day or cover the normal school day . . . and cover the normal school year” (Kamerman & Gatenio, 2003, pp. 1–2). Preschools share an educational component, but it is difficult to overstate their diversity. This policy arena includes a kaleidoscopic array of organi- zations and funding streams that has been labeled a “hodgepodge” (Kirp, 2007, p. 25) and a “mishmash” (Fuller, 2007, p. 285). One of the distinguishing features of preschool education in the United States is its public sector decentralization. Public preschool programs are administered at the national, state, and local levels.

The contemporary decentralization of preschool education in the United States gives tremendous discretion to state policymakers. At a fundamental level, they can choose to operate, or not to operate, publicly funded preschool programs within their jurisdictions. During the 2006–07 school year, 41 states dedicated public monies to preschool education (Rigby, 2007). This figure masks tremendous variation among the state-funded programs, which come in three basic forms. States may design and fund freestanding preschool programs, they may dedicate supplementary funds to the national Head Start program, or they may combine these two approaches. In 2006–07, 30 states funded freestanding programs, Oregon dedicated all of its pre- school spending to a Head Start supplement, and 10 states combined the two approaches.1

The contemporary political conflict over preschool education is multidimen- sional. In recent years, lawmakers in many states have devoted increased attention and funding to this policy arena. They have altered the eligibility requirements, curricular standards, personnel credentials, and classroom regulations that govern their state-funded programs. Such programmatic decisions are substantively signifi- cant because they determine who will gain access to state-funded preschool pro- grams and what type of services they will receive. Program eligibility has been an especially prominent topic, with policymakers and scholars debating the relative merits of universal versus targeted programs (Bushouse, 2009; Fuller, 2007; Gormley & Phillips, 2005; Kirp, 2007). This article focuses on the decision to devote public funds to preschool education, which precedes more specific programmatic choices and arguably has an even more profound impact on preschool availability and on the administration of such funds. The argument that “early learning programs merit a substantial commitment of public dollars” (Stone, 2008, p. 3) remains controversial, and state lawmakers have neither agreed on whether state funding is desirable nor settled on a single best way to administer such monies.

Why do states take divergent approaches to the funding of preschool education? This article examines the impact of several factors, some of which are especially relevant to preschool education and some of which reflect the general state political environment. It focuses on the role of the interest group community. A long-standing

218 Policy Studies Journal, 38:2

attribute of the preschool advocacy community is its internal division. Groups with the same underlying objectives, at least in theory, sometimes find themselves on opposite sides of a specific proposal or fighting a proposal that nominally serves these objectives. The next section examines the evolution of this internal division and explains why it might influence the contemporary funding of preschool education in the American states.

The Evolution of a Divided Constituency

In the United States, there has been a long-standing debate about the best way to educate and care for very young children and about the appropriate governmental role in this policy arena. This broad debate dates back at least to the infant school movement of the early 1800s (Beatty, 1995; Cahan, 1989). It continued in the early twentieth century, epitomized by the distinction between day nurseries and nursery schools. The former were generally perceived as a charitable enterprise serving low-income children whose mothers were forced to work, while the latter empha- sized educational development and maintained strong links to middle-class children and their parents (Michel, 1999; Rose, 1999). Governmental involvement in the education and care of very young children took on increased political salience during the Great Depression and the Second World War. Publicly funded “emer- gency” nursery schools were one component of the New Deal, and the national government created a targeted program, providing grants for day care to those areas most affected by the wartime mobilization effort. Preschool supporters hoped that these programs would lay the foundation for permanent government involvement in early childhood education and care, but these government policies were generally perceived to be temporary responses to crisis situations (Beatty, 1995; Lazerson, 1972; Slobdin, 1975).

During the postwar period, several developments altered the basic contours of the broader debate over the education and care of very young children. The large- scale entry of mothers with young children into the workforce altered social atti- tudes toward female employment and the governmental role in day care provision. As the employment rate of married women with children under six increased dra- matically, Americans increasingly viewed day care as a “socially legitimate ‘need’ of normal families, and even a potential responsibility of the state” (Rose, 1999, p. 213). At approximately the same time, research in developmental psychology increased scholarly and popular interest in early childhood education. Benjamin Bloom and J. McVicker Hunt challenged the prevailing wisdom that intelligence was fixed at birth and argued that carefully controlled stimulation was crucial to children’s mental, physical, and emotional growth during the early years of life (Bloom, 1964; Hunt, 1961). Preschool education gained heightened political salience as a result of these demographic and intellectual forces.

One manifestation of the increased interest in early childhood education and care was the establishment of Head Start as part of the War on Poverty in 1965. Head Start is a targeted national program that distributes federal funds to community organizations. It serves disadvantaged young children and their families (for the

Karch: Policy Feedback and Preschool Funding 219

most part), and epitomizes a comprehensive approach to child development. Head Start centers provide health care, nutritional, and other social services to the families while also preparing the children for primary school. Head Start maintained a tenuous existence during its early years, due in part to concerns about its long-term educational impact, but supporters managed to put it on more secure footing during the 1970s (Vinovskis, 2005; Zigler & Muenchow, 1992). The creation and political solidification of Head Start marked a key turning point in the development of American preschool education.

The adoption of a public policy can facilitate the organization and empowerment of its beneficiaries, a phenomenon known as policy feedback (Pierson, 1993, 2004). Head Start generated policy feedback, as its beneficiaries pressed for its expansion and sought to protect it against attack. Beneficiaries and supporters of Head Start, with backing from the African-American community and liberal Democrats, tend to be staunch defenders of the program. Head Start parents, for example, value Head Start’s educational and other services, the decision-making authority the program allows them to exercise, and the job opportunities it provides. Head Start families are “an integral part of the decision-making process” in governing and planning the program (Grotberg, 1981, p. 12). Although parents do not make determinative deci- sions on such issues as curriculum and budgeting, parent policy councils give them a voice in those decisions. Head Start also provides parents with career development opportunities that historically have focused on careers in early childhood care and education (Zigler & Muenchow, 1992, p. 118). One “side-effect” of the parent involve- ment component of Head Start was the emergence of the National Head Start Association, a national association of parents, directors, staff, and “friends” of Head Start that formed in 1973 (Zigler & Muenchow, 1992, p. 120). As a result, Head Start has been described as “a highly participatory, capacity-building program for mothers of low-income children,” and there is evidence that experience with the program has a positive effect on political participation (Schneider & Sidney, 2009, p. 110).

The political mobilization and empowerment of Head Start supporters, ironi- cally, contributed to the further fragmentation of the early education policy commu- nity during the late twentieth century. While Head Start supporters generally favored the expansion of publicly funded preschool programs at the state and national level, they were wary of how potential changes would affect their existing prerogatives. Preschool advocates divided into competing camps, and they engaged in a debate over teaching philosophies, program content, and parental responsibilities that manifested itself in turf battles over administrative control. As Congress debated the proper role of the national government in preschool education during the early and mid-1970s, one such battle erupted over the question of who should serve as a “prime sponsor” under an expanded governmental program. As part of a broader coalition, Head Start supporters argued that community-based organizations should play the preeminent administrative role. They were opposed by a second camp, supported mainly by teacher unions and the educational establishment, who favored the creation of a national preschool system under the purview of the nation’s public schools. Both camps favored an expanded governmental role in the provision of early childhood education and care, yet their competing policy prescriptions caused them

220 Policy Studies Journal, 38:2

to work at cross purposes. The sponsorship question helped derail such proposals as the Comprehensive Child Development Act of 1971 and the Child and Family Services Act of 1975 (McCathren, 1981; Zigler & Muenchow, 1992).

The issue of administrative control reemerged when the American Federation of Teachers (AFT) voiced its support for a national preschool system in the mid- 1970s. The union argued that the public schools should serve as the prime sponsor of such a system. As a result, many early childhood educators viewed the AFT proposal as little more than a power grab. In response, they voiced their support for a pluralistic approach, including multiple services offered by a wide variety of sponsors. They argued, “Parents will want different things for their children, and options should be available to enable them to make meaningful choices” (Fishhaut & Pastor, 1977, p. 47). Thus, the AFT proposal, rather than strengthening the coa- lition supporting public preschool provision, exacerbated existing tensions among preschool supporters. It illustrates the tension between the educational establish- ment and the community-based organizations that provided preschool services through such programs as Head Start. This tension contributed to the demise of the Child Care Act of 1979, a modest proposal that tried to foster a compromise among the various interest groups with a stake in early childhood education and care (McCathren, 1981).

The early education policy community remained internally divided in the 1980s. The reaction of the National Head Start Association to Smart Start, a bill introduced by Massachusetts Senator Edward M. Kennedy in 1988, is illustrative. Smart Start sought to provide health care, nutritional, social, and educational services to all 4-year-olds in the United States. During testimony before the Senate Committee on Labor and Human Resources, the president of the National Head Start Association voiced her misgivings about the bill. She first lobbied for “full and expanded support” for Head Start and then expressed her fear that the enactment of Smart Start would establish two funding streams and that the Congress would face pressure to combine them. She stated that the “solution of taking money from Head Start and giving it to Smart Start is exactly what we fear.”2 Her testimony suggests a continu- ing tension within the early education policy community as beneficiaries of pro- grams like Head Start viewed the creation of a freestanding preschool program as something of a threat. The adoption and political solidification of Head Start encour- aged its beneficiaries and supporters to mobilize politically in order to defend their existing prerogatives.

Divisions within the early education policy community may explain why states presently take divergent approaches to the funding of preschool education. The episodes profiled above illustrate how different constituencies, while nominally sharing the goal of increased public investment in preschool, offer divergent views of how such funds should be spent. In addition to the philosophical differences that prevent members of the early education policy community from seeing eye to eye, the public sector decentralization of preschool education seems to be an obstacle to cooperation. The adoption of Head Start and the mobilization and empowerment of its beneficiaries and supporters illustrate how policy feedback can influence interest group politics.

Karch: Policy Feedback and Preschool Funding 221

In recent years, the most significant policy breakthroughs in preschool education have occurred at the state level, as publicly funded programs have continued to emerge and expand. Case studies of these state-level developments suggest that divisions within the early education policy community and the existence of Head Start have affected the evolution of this policy sector. The Head Start community in Oklahoma, for example, was “initially suspicious” of plans to make that state’s preschool program universal but eventually supported the expansion after being convinced that it would increase revenues for Head Start and help achieve their other policy objectives (Bushouse, 2009, p. 43). Even so, Head Start leaders in Oklahoma expressed “growing concern about the possibility of a two-tiered preschool system” in the state (Fuller, 2007, p. 136). In West Virginia, the main political opposition to the creation of a universal preschool program came from a state representative who feared that “state funding would replace federal Head Start funding for preschool services,” an outcome that had already occurred in several counties within the state (Bushouse, 2009, p. 48). Similarly, changes in early childhood education in Texas were complicated by the fact that the public schools, Head Start centers, and childcare centers rarely conferred with one another and, in fact, “viewed one another with disdain” (Kirp, 2007, p. 198). Finally, Head Start leaders in Tennessee initially opposed the creation of a universal preschool program based on its teacher certification requirements because they believed such requirements would have a negative effect on Head Start enroll- ments (Bushouse, 2009, p. 129). These and other case studies of recent developments at the state level suggest that accommodating the preferences of the Head Start community is frequently a prerequisite for major changes in early education policy. A primary limitation of these analyses, however, is their focus on successful efforts at policy change. Almost without exception, they examine states in which the creation of a universal preschool program or the expansion of an existing program occurred. Understanding the political sources of preschool education policy requires attention to successful and unsuccessful attempts to change the status quo.

This article therefore extends the case study research described in the preceding paragraph through a quantitative analysis of contemporary preschool education across all 50 states. The preceding historical episodes and state-level case studies suggest that the Head Start community may represent a hurdle to the adoption of a freestanding, publicly funded preschool program at the state level. States in which Head Start supporters are a powerful political force may therefore be less likely to fund preschool education. As a result, the differences in preschool funding regimes described at the outset of this article may be because of the relative strength of the Head Start community at the state level. The next section describes an empirical assessment of this potential relationship.

Assessing the Determinants of Preschool Funding Strategies

In order to determine whether and how political factors affect the public funding of preschool education in the American states, this article examines two outcomes of interest, both of which are based on the preschool spending data developed by Rigby (2007). Government spending on preschool education falls into

222 Policy Studies Journal, 38:2

two categories: spending on freestanding preschool programs and spending on Head Start supplements. States therefore fall into one of four categories. The first category consists of states that do not dedicate any public funds to preschool edu- cation. The second category consists of states whose preschool education spending goes exclusively to a Head Start supplement. The third category consists of states that spend public monies both on a freestanding state preschool program and on a Head Start supplement. The final category consists of states that dedicate public funds exclusively to a freestanding preschool program.

The historical tension between Head Start supporters and other members of the early childhood policy community suggests that interest group politics are most likely to affect whether states fall into the first or the fourth category.3 If Head Start supporters in a state view the adoption of a freestanding preschool program as a political threat, then they may press state lawmakers not to create such a program. The congressional history of preschool education suggests that such a dynamic is possible. As a result, states with a large and politically engaged Head Start commu- nity may be especially likely not to fund preschool education. At the same time, they may be especially unlikely to dedicate all of their preschool funding to a freestanding state program.4 This article therefore focuses on two of the categories described in the preceding paragraph. The first dependent variable on which it focuses is a dichoto- mous variable indicating whether a state does not dedicate any public funds to preschool education. A positive relationship between the strength of the Head Start community and this outcome is expected. The second dependent variable is a dichotomous variable indicating whether a state dedicates all of its public spending on preschool education to a freestanding program. A negative relationship between the strength of the Head Start community and this outcome is expected.

The strength of the Head Start community is the key independent variable in the analysis that follows. The proxy used to assess its strength is the percentage of preschool attendees in a state who are enrolled in Head Start. This proxy is probably best described as a measure of the potential political strength of the Head Start community; it describes the size of this community and, thus, cannot account for its political activity. State Head Start organizations possess divergent opinions on such issues as the value of expanding preschool access, and the political effectiveness of their leaders may also vary. The chief limitation of the proxy employed here is that it does not account for these differences in policy stances and leadership. The main advantage of this proxy is that it represents an objective, unbiased measure of the reach of Head Start relative to other programs of early childhood education in a particular state. Data on Head Start enrollment at the state level come from the Administration for Children and Families of the Department of Health and Human Services.5 Data on overall preschool enrollment at the state level come from the U.S. Census Bureau.6 The Head Start percentage ranges from a low of 9.2 percent in New Hampshire and New Jersey to a high of 65.3 percent in Mississippi. States in which a high percentage of preschool enrollees are enrolled in Head Start are expected to be especially likely not to fund preschool education and especially unlikely to dedicate all preschool spending to a freestanding state program. These expectations are based on the historical episodes and case study evidence profiled in the preceding section,

Karch: Policy Feedback and Preschool Funding 223

which suggest that the Head Start community is often reluctant, at least initially, to support the creation or expansion of state-funded preschool programs. Presumably, state policymakers will be more attentive to the preferences of the Head Start com- munity when that community represents a larger proportion of those with an exist- ing stake in preschool education.

In addition to the size of the Head Start community, it is necessary to consider the potential impact of other correlates of state policymaking. Many features of a state’s demographic and political environment might affect decisions about pre- school education funding. One potential demographic factor is the age profile of the state population. The issue of preschool education might resonate more strongly in states with a relatively high proportion of young children. Officials may perceive a greater need for publicly funded preschool programs in such states. The logic of this potential relationship is analogous to the argument that policy adoption is more likely when the policy responds to conditions within a state (Nice, 1994). The analysis that follows includes the proportion of the state population that is under 5 years old.7

States with a younger population base might face stronger demand for preschool services and, therefore, might be especially likely to fund such services.8

State wealth is a second demographic trait with the potential to affect preschool funding decisions. Its potential impact is related both to the specific population served by Head Start and the general impact of state wealth on the making of public policy. Head Start eligibility is based on income levels and poverty guidelines, with the vast majority of enrollees living in low-income families. The correlation between state income levels and Head Start enrollment is not perfect because the program consistently does not reach the entire eligible population. The inclusion of a proxy for state wealth therefore insures that a significant relationship between the strength of the state Head Start community and preschool funding decisions is not merely an outgrowth of the socioeconomic profile of a state. In addition, relatively wealthy states may have more resources available to spend on preschool education, while less wealthy states may be reluctant to devote their scarcer resources to this policy arena. Political scientists have long noted a general relationship between state wealth and policy adoption (Dye, 1966; Walker, 1969), a relationship which implies that policy adoption is driven by “resources rather than needs” (Tweedie, 1994). The analysis that follows uses real state per capita income as its proxy for the economic resources that might facilitate preschool funding.9

Other general demographic attributes of a state might also affect the decision to dedicate government funds to preschool education. The average formal education level of the state population might affect societal attitudes toward preschool and, by extension, the willingness to fund it. States whose residents have a higher average level of formal education might be more sympathetic to public spending on educa- tion in general and on preschool education in particular. The analysis that follows uses the percentage of the state population with a high school diploma or higher as its proxy for the education level of the state.10 A positive relationship between state education level and state preschool spending is expected. Another potential demo- graphic influence is the racial diversity of the state. Scholars have linked state racial heterogeneity to state welfare policy choices (Fellowes & Rowe, 2004; Soss, Schram,

224 Policy Studies Journal, 38:2

Vartanian, & O’Brien, 2001), and racial diversity has also been linked to educational outcomes at the state level (Hero & Tolbert, 1996). In general, these studies find that states with larger minority populations tend to have less generous social policies. To investigate whether a similar relationship holds in the context of preschool educa- tion, the analysis that follows includes the percentage of the state population iden- tified as non-Hispanic white as a proxy for homogeneity.11 A positive relationship between population homogeneity and state preschool spending is expected.

The political environment might also influence decisions about preschool funding at the state level. Specifically, party control of government institutions (Roh & Haider-Markel, 2003; Spill, Licari, & Ray, 2001) and the ideological environment (Berry, Ringquist, Fording, & Hanson, 1998; Erikson, Wright, & McIver, 1993) have been linked to the enactment of various public policies. The two major political parties tend to possess distinct policy priorities, and Democrats are usually more enthusiastic about publicly funded preschool education than Republicans are. Democratic control of state political institutions might therefore increase the likeli- hood that the state will devote public funds to preschool education, while Republi- can control of the governorship or the state legislature might make such funding less likely. The analysis that follows therefore includes a dichotomous variable that indi- cates whether Democrats exercise unified party control of state government.12 A similar logic explains the potential relationship between preschool funding and the ideological environment. While conservatives generally prefer public policies that reduce the scope of government activity, liberals typically support the expansion of government prerogatives. Many conservatives characterize publicly funded pre- school education as an undesirable intrusion on parental autonomy. As a result, states in which residents hold relatively liberal political views might be more likely to dedicate government funds to preschool education. The analysis that follows there- fore includes an annual estimate of citizen liberalism (Berry et al., 1998). A positive relationship between preschool funding and both democratic control and citizen liberalism is expected.

To evaluate the potential relationships discussed in the preceding paragraphs, this article examines 5 years of cross-sectional state preschool spending data devel- oped by Rigby (2007). The data cover the period from 2001 through 2005; the unit of analysis is a state year. The two outcomes of interest, whether a state chose not to dedicate public funds to preschool and whether a state dedicated all preschool funding to a freestanding state program, are investigated in two separate models. Each model includes all of the independent variables described in this section in addition to indicator variables for each year of the analysis. Because both of the dependent variables are dichotomous, standard logistic regression methods are used to evaluate the determinants of preschool funding decisions. The next section describes the results of this empirical analysis.

Results

Why do some states choose not to dedicate public funds to preschool education? Table 1 displays the results of a logistic regression model examining the determi-

Karch: Policy Feedback and Preschool Funding 225

nants of this decision. The “first differences” displayed in the table (and in Table 2) are predicted probabilities rather than the coefficients of the initial analysis. They were derived by manipulating the quantity of interest and setting all continuous variables to their means and all dichotomous variables to their modes. For such variables as the size of the Head Start community, the values in the tables display the change in the predicted probability when the quantity of interest shifts from one standard deviation below its mean to one standard deviation above its mean. For dichotomous variables like unified Democratic government, the values in the tables display the change in predicted probability when the quantity of interest shifts from zero to one.13

As expected, interest group politics, specifically the division within the early childhood policy community, is influential. The size of the Head Start community has a positive and significant effect on the likelihood that a state will not allocate public funds to preschool. All else equal, moving from a state with a relatively small Head Start community (12.95 percent of total preschool enrollment) to a state with a relatively large one (35.30 percent of total preschool enrollment) increases the like- lihood that a state will not fund preschool by 12.72 percentage points. This result suggests that the Head Start community perceives other publicly funded preschool programs as a threat and that state officials are responsive to the preferences of this constituency when it represents a fairly large proportion of those with an existing stake in preschool education.

Other demographic and political factors also appear to affect the decision not to fund preschool education. As predicted, there is a negative and significant relation- ship between state wealth and this decision. Wealthy states are significantly more

Table 1. Determinants of the Decision Not to Fund Preschool Education

Variable First Difference

Confidence Interval

Head Start enrollment (+) 12.72** 3.12, 30.31 Population under five (-) 0.71 -5.26, 7.88 Per capita income (-) -12.57* -31.53, -2.02 Education level (-) 15.20** 4.54, 33.01 Population homogeneity (-) 2.71 -4.82, 14.40 Democratic government (-) -3.15 -9.05, -0.14 Citizen ideology (-) -6.53** -17.21, -1.17 2002 -0.49 -6.42, 6.15 2003 0.88 -6.13, 8.33 2004 5.61 -3.13, 18.47 2005 4.82 -3.26, 16.13 Number of observations 250 Log likelihood -65.368 Chi-square 98.79 Prob. chi-square 0.000 Pseudo r-squared 0.430

Note: Expected directions in parentheses. All tests of statistical signifi- cance are two-tailed. *Significant at the 0.05 level; **significant at the 0.01 level.

226 Policy Studies Journal, 38:2

likely to allocate public funds to preschool.14 All else equal, moving from a relatively wealthy state to a relatively poor one decreases the likelihood that a state will fund preschool by 12.57 percentage points. This result suggests that resource availabil- ity facilitates public funding of preschool education. In addition, citizen ideology has the expected negative and significant relationship with this particular funding decision. As states become more liberal, they become increasingly likely to fund preschool. All else equal, moving from a relatively liberal state to a relatively conservative state decreases the likelihood that a state will offer publicly funded preschool by 6.53 percentage points. The presence of unified Democratic government in a state also appears to increase the likelihood that it will fund preschool education, but this relationship does not achieve conventional levels of statistical significance. The effects of the remaining three variables included in the model—percentage of the state population under age five (positive), state education level (positive), and popu- lation homogeneity (positive)—do not conform to the predictions of the preceding section. The relationship between the state education level and the decision not to fund preschool education is the only one of these three that achieves conventional levels of statistical significance.

Table 2 displays the results of a model investigating the determinants of the decision to devote public funds exclusively to a freestanding state preschool program. As predicted, the division within the early childhood policy community affects the likelihood that a state will follow this funding strategy. The size of the Head Start community has a negative and significant effect. All else equal, moving from a state with a relatively small Head Start community to a state with a relatively large one reduces the likelihood that a state will exclusively fund a freestanding

Table 2. Determinants of the Decision to Fund Only a Freestanding Preschool Program

Variable First Difference

Confidence Interval

Head Start enrollment (-) -43.25** -61.90, -23.32 Population under five (+) -5.48 -25.95, 15.49 Per capita income (+) -15.47 -39.27, 7.32 Education level (+) -50.21** -65.52, -32.60 Population homogeneity (+) -3.97 -21.87, 15.08 Democratic government (+) 24.50* 2.18, 43.45 Citizen ideology (+) 3.23 -17.16, 21.65 2002 4.57 -17.20, 27.60 2003 -0.34 -22.67, 22.61 2004 21.82 -0.68, 43.31 2005 24.48* 2.87, 46.09 Number of observations 250 Log likelihood -132.611 Chi-square 75.55 Prob. chi-square 0.000 Pseudo r-squared 0.222

Note: Expected directions in parentheses. All tests of statistical signifi- cance are two-tailed. *Significant at the 0.05 level; **significant at the 0.01 level.

Karch: Policy Feedback and Preschool Funding 227

preschool program by 43.25 percentage points. Like the analogous result in Table 1, this result suggests that the Head Start community views a freestanding preschool program as a threat and that state officials are responsive to the preferences of this constituency when it represents a relatively large proportion of those with an exist- ing stake in preschool education. Interest group politics, as anticipated, seems to affect state decisions about whether and how to fund preschool education.

The performance of the other demographic and political factors examined in the model is somewhat disappointing. As expected, unified Democratic govern- ment makes it more likely that a state will fund only a freestanding preschool program. All else equal, moving from a state without unified Democratic govern- ment to a state with that partisan alignment increases the probability that the state will fund only a freestanding preschool program by 24.50 percentage points. Citizen ideology also has the anticipated effect, as liberal states are more likely to follow this specific funding strategy. This relationship, however, does not achieve conventional levels of statistical significance. The effects of the remaining four vari- ables included in the model—percentage of the population under five (negative), state wealth (negative), state education level (negative), and population homoge- neity (negative)—do not conform to the predictions of the previous section. The relationship between the education level and the exclusive funding of a freestand- ing state program is the only one of these four that achieves conventional levels of statistical significance.

The results presented in Tables 1 and 2 suggest that the size of the Head Start community affects whether and how states fund preschool education.15 A relatively large Head Start presence increases the likelihood that a state will not dedicate funds to preschool and decreases the likelihood that a state will dedicate all of its public preschool funds to a freestanding program.16 State wealth appears to affect whether a state chooses to fund preschool but not the specific funding strategy chosen. The political environment also appears to be influential, as relatively liberal states are more likely to fund preschool education, and the existence of unified Democratic government increases the likelihood that a state will fund only a freestanding preschool program. In contrast, neither the age profile of a state nor its racial and ethnic diversity appears to have a significant impact on these two decisions. The relationship between the state education level and preschool funding decisions presents the most puzzling results in Tables 1 and 2. Although state education levels appear to influence both of the outcomes examined in this article, their impact is an unexpected direction. The negative relationship between education and preschool funding might reflect regional patterns in this policy arena. Although southern states generally have populations with less formal edu- cation, such states as Georgia and Oklahoma have been leaders in preschool invest- ment. This southern leadership dates to the 1980s, when such regional groups as the Commission on the Future of the South and the Southern Regional Education Board called for increased investment in public schools, including preschool pro- grams, as part of a long-term economic growth strategy and a way to make the South more attractive to the business community and to professional workers (Bushouse, 2009, p. 19).

228 Policy Studies Journal, 38:2

Conclusion

Preschool education in the United States has long been a sector characterized by decentralization and diversity, with many critical policy decisions made at the state level. During the 2006–07 school year, 41 states dedicated public monies to preschool education, and state spending on these programs has increased by hundreds of millions of dollars in recent years (Rigby, 2007). The preceding analysis suggests that interest group politics affects both the decision to fund preschool education and the decision to operate a freestanding preschool program. These decisions appear to reflect a division within the early education policy community, with Head Start playing an especially important role. While Head Start supporters usually favor the expansion of publicly funded preschool, they tend to be skeptical of proposals that seem to affect their existing prerogatives. At the congressional level, they have occasionally portrayed such proposals as a threat to the long-term existence of Head Start. Case studies of recent developments at the state level suggest that accommo- dating the preferences of the Head Start community is frequently required in order to achieve major policy change in preschool education. The preceding analysis sug- gests that states with a large Head Start community are significantly more likely not to fund preschool education and significantly less likely to dedicate preschool funding exclusively to a freestanding state program.

The impact of the Head Start community is consistent with a phenomenon known as policy feedback. Policy feedback suggests the existing policy repertoire has a dynamic impact on policymaking, serving as a “vital force shaping the alter- natives perceived and the policies adopted” (Heclo, 1974, p. 156). Officials who wish to alter the status quo must sometimes overcome the opposition of constituencies who benefit from existing policy arrangements, and the political power of program beneficiaries can become a source of stability that constrains the options that poli- cymakers possess. Once public policies are established, they affect the possibilities for future policymaking. Existing policies shape the identities, interests, and incen- tives of key social actors (Skocpol, 1992). In this sense, the political dynamics of early childhood education and care resonate with congressional debates over such issues as retirement security and health care where the mobilization of existing stakehold- ers has been an obstacle to comprehensive policy change (Campbell, 2003; Hacker, 2002; Mayes, 2004; Walker, 1991).

The chief limitation of the quantitative analysis presented in the preceding section is that it cannot isolate the precise mechanism or mechanisms through which policy feedback occurs in preschool education. It cannot shed light on the specific actions taken by Head Start supporters as they defend the program. The results presented in this article may be because of the political activity of Head Start leaders, the worries about the financial implications of policy change for Head Start and the state more generally, or a feeling among state political leaders that large Head Start enrollments obviate the need for state efforts to educate and care for very young children. Case studies that compare successful and unsuccessful attempts at major policy change in preschool education will add greater depth to the results presented in this article. Another potential avenue for future research is to extend the present

Karch: Policy Feedback and Preschool Funding 229

study to other dimensions of preschool education policy, such as curricular stan- dards, personnel credentials, and classroom regulations. The contemporary political debate over preschool education is multidimensional, and the size of the Head Start community may or may not be relevant in other contexts.

The fragmentation of the advocacy community seems likely to affect the ongoing evolution of preschool education at the state level. Some observers have already noted the difficulty of integrating Head Start into universal state programs serving all children (Barnett & Hustedt, 2003, p. 56), although others have pointed to suc- cessful examples of such integration (Gormley, Phillips, & Gayer, 2008). In recent years, the issue of coordination among preschool providers has received heightened attention during congressional hearings. In testimony before a subcommittee in the U.S. House, the executive director of the Ohio Head Start Association characterized “collaboration versus competition with state funded [preschool] programs” as one of the many new challenges that local Head Start programs will face in the coming years.17 The historical development of preschool education in the United States and case studies of recent adjustments to state preschool programs suggest that accom- modating existing stakeholders is a prerequisite for successful reform efforts (Bush- ouse, 2009; Fuller, 2007; Kirp, 2007). Collaboration between state prekindergarten programs and Head Start is certainly possible, but, as the interest group dynamics depicted in this article seem to suggest, it will not be a foregone conclusion.

Andrew Karch is an associate professor in the Department of Government at the University of Texas at Austin.

Notes

I thank the anonymous reviewers, Elizabeth Rigby, Jal Mehta, Stephanie Rubin, Dick Winters, Nancy Martorano Miller, and Scott Abernathy for their constructive comments. I also thank Elizabeth Rigby for sharing some of the data used in my analysis.

1. See Rigby (2007) for a thorough description of the data on which these figures are based.

2. Senate Committee on Labor and Human Resources, Smart Start: The Community Collaboration for Early Childhood Development Act of 1988: Hearings on S. 2270, 100th Cong., 2d. sess., 1988, pp. 294–95.

3. One might also hypothesize that states with a relatively strong Head Start community will be more likely to dedicate public funds exclusively to a Head Start supplement. This article does not evaluate that specific hypothesis, however, because very few states undertake such a funding approach.

4. The relationship between a strong Head Start community and a mixed funding approach is unclear. Head Start supporters may favor the supplemental funding dedicated to their preferred program but nonetheless worry that the simultaneous existence of a freestanding program represents a long-term threat. In addition, Head Start supplements sometimes come with “strings attached.” While some supplements fund extended days or minimal enrollment expansions, in other cases, states assume some administrative responsibility for the Head Start program (Barnett, Hustedt, Hawkinson, & Robin, 2006, p. 31). As a result, the Head Start community may actually resist the establishment of a mixed funding approach.

5. Administration for Children and Families. “Head Start Program Fact Sheet: Fiscal Year 2002.” http:// www.acf.hhs.gov/programs/ohs/about/fy2002.html. Accessed March 12, 2009.

230 Policy Studies Journal, 38:2

6. U.S. Census Bureau. Census 2000 Supplementary Survey, “Table P053: School Enrollment by Level of School for the Population 3 Years and Over.”

7. U.S. Census Bureau. “SC-EST2005-02: Population Estimates by State: Age and Sex for States and for Puerto Rico: April 1, 2000 to July 1, 2006.” http://www.census.gov/popest/states/asrh/SC-EST2005- 02.html. Accessed February 27, 2008.

8. When this section refers to anticipated “positive” or “negative” relationships between public spend- ing on preschool education and various state demographic and political characteristics, it is important to recognize that the two dependent variables included in the analysis that follows actually complicate such a statement. The first dependent variable indicates whether a state does not fund preschool education. As a result, any characteristic that is expected to have a “positive” relationship with preschool spending will be expected to have a negative relationship with this specific outcome. The second dependent variable indicates whether a state dedicates all of its preschool funding to a freestanding program. The expected relationships discussed in this section are therefore expected to hold as they are described here. For ease of interpretation, Tables 1 and 2 in the next section will include the expected direction of the relationship between the various factors and the specific out- comes under examination.

9. The measure used is per capita income (Bureau of Economic Analysis, http://www.bea.gov/ regional/reis, accessed February 20, 2008) divided by the implicit price deflator (“Table D-1: Popula- tion, U.S. Gross Domestic Product, and Implicit Price Deflator,” http://www.eia.doe.gov/emeu/aer/ txt/ptb1601.html, accessed February 20, 2008) and then logged.

10. The measure is a percentage of the state population age 25 and over. U.S. Census Bureau. Census 2000 Summary File 3 (SF-3) Sample Data, “QT-P20. Educational Attainment by Sex: 2000.”

11. U.S. Census Bureau, “SC-EST2006-6RACE: Annual State Population Estimates with Sex, 6 Race Groups (5 Race Alone Groups and One Group with Two or More Race Groups) and Hispanic Origin: April 1, 2000 to July 1, 2006.” http://www.census.gov/popest/states/asrh/stasrh.html. Accessed March 11, 2010.

12. The party control measure is an updated, annual version of the data presented in Klarner (2003).

13. The values were derived based on the statistical simulation technique and computer software described in King, Tomz, and Wittenberg (2000). The “confidence intervals” presented in the tables are 2.5 percent, 97.5 percent of the posterior distributions.

14. Technically, the results presented in Table 1 indicate that relatively wealthy states, relatively liberal states, and states with unified Democratic government are all significantly less likely not to allocate public funds to preschool. The formulations used in the text have been employed for ease of interpretation.

15. The key relationship identified in this article is robust to changes in model specification. In models that substitute three distinct indicator variables for Democratic governor, upper chamber, and lower chamber for unified Democratic control, which replace the measure of population homogeneity with the percentage of the population identified as “Black” and “Latino,” and that incorporate the percent- age of the population with a bachelor’s degree rather than the percentage with a high school diploma, the relationship between the size of the Head Start community and preschool funding is in the expected direction and achieves conventional levels of statistical significance. These models suggest that neither the absence of a relationship between ethnic and racial diversity and preschool funding decisions nor the unexpected relationship between state education levels and preschool funding decisions is an artifact of the proxies used to measure these state attributes. They also suggest that states in which Democrats control the upper chamber are more likely to fund preschool education and states with Democratic governors are more likely to operate a freestanding preschool program. In another test of the robustness of the results reported in the present section, the analysis was also performed using a random effects model. The substantive results of the analysis hold, as the interest group variable achieves conventional levels of statistical significance in both models. Using a random effects model does not affect the results reported in Table 2, as an identical set of variables achieves conventional levels of statistical significance. Using a random effects model has a slight impact on the results reported in Table 1. State wealth does not achieve conventional levels of statistical significance; the proportion of the state population under age five, however, has an unanticipated positive and significant effect on the likelihood that a state will not fund preschool education. In light of the main

Karch: Policy Feedback and Preschool Funding 231

theme of this article, the most important result of this auxiliary test is the robustness of the relationship between the interest group variable and the two main outcomes of interest.

16. Of course, Head Start beneficiaries and supporters are not the only constituency with a potential interest in maintaining the status quo. Preschool and childcare providers in the private sector might also possess a similar interest. To examine the potential impact of private sector providers, alternate models included two different proxies of private sector enrollment. This variable is not included in the models presented here because the two proxies produced conflicting results. Importantly, however, the inclusion of either proxy did not affect the relationship between the size of the Head Start community and preschool funding, which remained in the expected direction and attained conven- tional levels of statistical significance in every case. Teacher unions represent another potential constituency with a stake in this policy arena, and including a recent measure of unionization rates also did not meaningfully affect the primary relationship in which this article is interested.

17. House Subcommittee on Early Childhood, Elementary and Secondary Education of the Committee on Education and Labor. Improving Head Start for America’s Children, 110th Cong., 1st sess., 2007, p. 23.

References

Barnett, W. Steven, and Jason T. Hustedt. 2003. “Preschool: The Most Important Grade.” Educational Leadership 60: 54–57.

Barnett, W. Steven, Jason T. Hustedt, Laura E. Hawkinson, and Kenneth B. Robin. 2006. The State of Preschool: 2006 State Preschool Yearbook. New Brunswick, NJ: National Institute on Early Education Research.

Beatty, Barbara. 1995. Preschool Education in America: The Culture of Young Children from the Colonial Era to the Present. New Haven, CT: Yale University Press.

Berry, William D., Evan J. Ringquist, Richard C. Fording, and Russell L. Hanson. 1998. “Measuring Citizen and Government Ideology in the American States, 1960–1993.” American Journal of Political Science 42: 327–48.

Bloom, Benjamin. 1964. Stability and Change in Human Characteristics. New York: Wiley.

Bushouse, Brenda K. 2009. Universal Preschool: Policy Change, Stability, and the Pew Charitable Trusts. Albany, NY: SUNY Press.

Cahan, Emily D. 1989. Past Caring: A History of U.S. Preschool Care and Education for the Poor, 1820–1965. New York: National Center for Children in Poverty.

Campbell, Andrea. 2003. “Participatory Reactions to Policy Threats: Senior Citizens and the Defense of Social Security and Medicare.” Political Behavior 25: 29–49.

Dye, Thomas R. 1966. Politics, Economics, and the Public: Policy Outcomes in the American States. Chicago: Rand McNally.

Erikson, Robert, Gerald C. Wright, and John P. McIver. 1993. Statehouse Democracy: Public Opinion and Policy in the States. Cambridge, UK: Cambridge University Press.

Fellowes, Matthew C., and Gretchen Rowe. 2004. “Politics and the New American Welfare States.” American Journal of Political Science 48: 362–73.

Fishhaut, Erna H., and Donald Pastor. 1977. “Should the Public Schools be Entrusted with Preschool Education: A Critique of the AFT Proposals.” School Review 86: 38–49.

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

Gormley, William T., Jr., and Deborah Phillips. 2005. “The Effects of Universal Pre-K in Oklahoma: Research Highlights and Policy Implications.” Policy Studies Journal 33: 65–82.

Gormley, William T., Jr., Deborah Phillips, and Ted Gayer. 2008. “Preschool Programs Can Boost School Readiness.” Science 320: 1723–4.

Grotberg, Edith H. 1981. “The Federal Role in Family Policies.” In Approaches to Child and Family Policy, ed. Harold C. Wallach. Boulder, CO: Westview Press, 9–24.

232 Policy Studies Journal, 38:2

Hacker, Jacob S. 2002. The Divided Welfare State: The Battle over Public and Private Social Benefits in the United States. Cambridge, UK: Cambridge University Press.

Heclo, Hugh. 1974. Modern Social Policy in Britain and Sweden: From Relief to Income Maintenance. New Haven, CT: Yale University Press.

Hero, Rodney E., and Caroline J. Tolbert. 1996. “A Racial/Ethnic Interpretation of Politics and Policy in the States of the U.S.” American Journal of Political Science 40: 851–71.

Hunt, Joseph McVicker. 1961. Intelligence and Experience. New York: Ronald Press.

Kamerman, Sheila, and Shirley Gatenio. 2003. “Overview of the Current Policy Context.” In Early Child- hood Education and Care in the USA, ed. Debby Cryer and Richard M. Clifford. Baltimore, MD: Paul H. Brookes, 1–30.

King, Gary, Michael Tomz, and Jason Wittenberg. 2000. “Making the Most of Statistical Analyses: Improv- ing Interpretation and Presentation.” American Journal of Political Science 44: 347–61.

Kirp, David L. 2007. The Sandbox Investment: The Preschool Movement and Kids-First Politics. Cambridge, MA: Harvard University Press.

Klarner, Carl E. 2003. “The Measurement of the Partisan Balance of State Government.” State Politics and Policy Quarterly 3: 309–19.

Lazerson, Marvin. 1972. “The Historical Antecedents of Early Childhood Education.” In Early Childhood Education, ed. Ira J. Gordon. Chicago: University of Chicago Press, 33–53.

Mayes, Rick. 2004. Universal Coverage: The Elusive Quest for National Health Insurance. Ann Arbor, MI: University of Michigan Press.

McCathren, Randall R. 1981. “The Demise of Federal Categorical Child Care Legislation: Lessons for the ‘80s from the Failures of the ‘70s.” In Approaches to Child and Family Policy, ed. Harold C. Wallach. Boulder, CO: Westview Press, 101–43.

Michel, Sonya. 1999. Children’s Interests/Mothers’ Rights. New Haven, CT: Yale University Press.

Nice, David C. 1994. Policy Innovation in State Government. Ames, IA: Iowa State University Press.

Pierson, Paul. 1993. “When Effect Becomes Cause: Policy Feedback and Political Change.” World Politics 45: 595–628.

———. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton, NJ: Princeton University Press.

Pierson, Paul, and Theda Skocpol. 2002. “Historical Institutionalism in Contemporary Political Science.” In Political Science: State of the Discipline, ed. Ira Katznelson and Helen V. Millner. New York: W. W. Norton, 693–721.

Rigby, Elizabeth. 2007. “Same Policy Area, Different Politics: How Characteristics of Policy Tools Alter the Determinants of Early Childhood Education Policy.” Policy Studies Journal 35: 653–70.

Roh, Jongho, and Donald P. Haider-Markel. 2003. “All Politics Is Not Local: National Forces in State Abortion Initiatives.” Social Science Quarterly 84: 15–31.

Rose, Elizabeth. 1999. A Mother’s Job: The History of Day Care, 1890–1960. New York: Oxford University Press.

Schneider, Anne, and Mara Sidney. 2009. “What Is Next for Policy Design and Social Construction Theory?” Policy Studies Journal 37: 103–19.

Skocpol, Theda. 1992. Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States. Cambridge, MA: Harvard University Press.

Slobdin, Carol. 1975. “When the U.S. Paid for Day Care.” Day Care and Early Education 3: 22–25, 49.

Soss, Joe, Sanford F. Schram, Thomas P. Vartanian, and Erin O’Brien. 2001. “Setting the Terms of Relief: Explaining State Policy Choices in the Devolution Revolution.” American Journal of Political Science 45: 378–95.

Spill, Rorie L., Michael J. Licari, and Leonard Ray. 2001. “Taking on Tobacco: Policy Entrepreneurship and the Tobacco Litigation.” Political Research Quarterly 54: 605–22.

Stone, Diana. 2008. Funding the Future: States’ Approaches to Pre-K Finance, 2008 Update. Washington, DC: Pre-K Now.

Karch: Policy Feedback and Preschool Funding 233

Tweedie, Jack. 1994. “Resources Rather than Needs: A State-Centered Model of Welfare Policymaking.” American Journal of Political Science 38: 651–72.

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

Walker, Jack L., Jr. 1969. “The Diffusion of Innovations among the American States.” American Political Science Review 63: 880–99.

———. 1991. Mobilizing Interest Groups in America: Patrons, Professions, and Social Movements. Ann Arbor, MI: University of Michigan Press.

Zigler, Edward, and Susan Muenchow. 1992. Head Start: The Inside Story of America’s Most Successful Educational Experiment. New York: Basic Books.

234 Policy Studies Journal, 38:2

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy.

Users should refer to the original published version of the material.