Report # 01-01
Office of the Legislative Auditor
State of Minnesota Program Evaluation Report

REPORT # 01-01

OFFICE OF THE LEGISLATIVE AUDITOR STATE OF MINNESOTA

PROGRAM EVALUATION REPORT Early Childhood Education Programs

JANUARY 2001

OLA

Centennial Building, 658 Cedar Street, St. Paul, Minnesota 55155 • Tel: 651/296-4708 • Fax: 651/296-4712 E-mail:legislative.auditor@state.mn.us • TDD Relay: 651/297-5353 • Website: www.auditor.leg.state.mn.us

OFFICE OF THE LEGISLATIVE AUDITOR STATE OF MINNESOTA • James Nobles, Legislative Auditor

January 12, 2001 Members Legislative Audit Commission

In April 2000, the Legislative Audit Commission directed us to study several of Minnesota’s early childhood education programs. Legislators asked questions about program funding and oversight, and they were interested in learning more about the people served by these programs and the services provided.

Although there are overlapping elements in Minnesota’s Head Start, Early Childhood Family Education (ECFE), and School Readiness programs, we found that these programs often differ considerably in the intensity and scope of services they provide. In addition, we identified various funding practices that should be reviewed by the Legislature. For instance, many school districts have accumulated large ECFE fund balances while receiving limited monitoring from the Department of Children, Families, and Learning, and state Head Start allocations could be better targeted to areas with unmet needs. Also, our evaluation found that there is limited knowledge about what types of early childhood programs are most effective, and there is a need for statewide measurement of participants’ subsequent performance in the K-12 school system.

This report was researched and written by Joel Alter (project manager), Judy Randall, and Leah Goldstein. We received the full cooperation of the Minnesota Department of Children, Families, and Learning, early childhood education service providers, and others.

Sincerely,

/s/ James Nobles /s/ Roger Brooks

James Nobles Roger Brooks Legislative Auditor Deputy Legislative Auditor

Table of Contents

SUMMARY ix INTRODUCTION 1 1. PROGRAM CHARACTERISTICS
Overview of Minnesota’s Early Childhood Service System Description of Minnesota’s Three Major Programs Differences Among Head Start, ECFE, and School Readiness Service Coordination
2. HEAD START FUNDING ISSUES
Program Availability and Participation Program Funding Impact of Recent Funding Increases Monitoring of Head Start Finances
3. ECFE AND SCHOOL READINESS FUNDING ISSUES
ECFE Program Availability and Participation ECFE Funding Issues School Readiness Program Availability and Participation School Readiness Funding Issues State Monitoring of District Finances
4. PROGRAM OUTCOMES AND OVERSIGHT General Research Findings Minnesota Studies of Program Outcomes Ongoing State Oversight
SUMMARY OF RECOMMENDATIONS FURTHER READING AGENCY RESPONSE 89 RECENT PROGRAM EVALUATIONS Back Cover

List of Tables and Figures

Page Tables

1.1 State and Federal Funding for Minnesota’s Early Childhood Programs, FY 2001 5 1.2 Selected Measures of Head Start Services in Minnesota and U.S.,

FY 1999 10 1.3 Enrollment in Minnesota Head Start Programs, FY 1999 11 1.4 ECFE Services, FY 1999 13 1.5 School Readiness Services, FY 1999 14 1.6 Participant Characteristics of Early Childhood Programs, FY 1999 17 1.7 Staff Characteristics of Early Childhood Programs, FY 1999 20 1.8 Program Differences Between Head Start, ECFE, and School

Readiness 22 2.1 Head Start Slots By County, as a Percentage of Children in

Poverty (Selected Counties) 29 2.2 Largest Cities Without Head Start Service Sites, 1999-2000 31 2.3 Minnesota Head Start Funding, FY 1988-2001 33 2.4 Head Start Innovative Grant Funding in Minnesota, FY 1991-2001 34 2.5 Changes in Minnesota Head Start Funded Slots and Contact Hours,

FY 1997-1999 40 3.1 ECFE Participation Rates, 1998-99, Selected School Districts 47 3.2 ECFE Per-Child Funding Allowance, FY 1986-2001 50 3.3 ECFE Revenues and Reserve Balances, FY 1999 51 3.4 ECFE Parent-Child Classes and Participants in Districts with 150

or Fewer ECFE-Eligible Children, FY 1999 52 3.5 Median Percentage of ECFE-Eligible Population Served, FY 1999 53 3.6 School Readiness Participation Rates, 1998-99, Selected School

Districts 57 3.7 School Readiness Funding, FY 1992-2001 60

4.1 Impacts of Preschool Programs--A Sampling of Research Summaries 68 Figures 2.1 Service Areas of Minnesota’s Regular Head Start Grantees 26 2.2 Head Start Slots as a Percentage of Children Under Five in Poverty 30

Summary Major Findings:

• Minnesota generously funds Head Start compared with other states, but the present method of allocating state Head Start funds throughout Minnesota does not adequately reflect the location of families in poverty or the cost of services (pp. 32, 37).

-Minnesota’s School Readiness program subsidizes preschool for children not served by Head Start, and funding for this program is modest compared with that in some states (p. 59).

-The Head Start, Early Childhood Family Education (ECFE), and School Readiness programs differ considerably in the services they provide and the intensity of these services, and this contributes to their widely differing costs (p. 15).

-Many school districts have large balances of unspent funds in their ECFE programs (p. 51).

-The Department of Children, Families, and Learning assists school districts with their ECFE and School Readiness programs, but its fiscal and on-site monitoring of local programs is minimal (pp. 62, 78).

-Studies of early childhood programs nationally have shown mixed results, and Minnesota studies have not provided definitive evidence of programrelated impacts (pp. 66, 74, 76).

Key Recommendations:

-The Legislature should consider increasing the percentage of state Head Start funds allocated on the basis of unmet need (p. 38) and increasing Head Start grantees’ flexibility to determine which families they serve (p. 42).

-The Legislature should consider options to discourage school districts from having large ECFE reserve balances (p. 53), and it should establish a School Readiness reserve fund (p. 61).

-The Department of Children, Families, and Learning should increase financial monitoring and on-site review of local ECFE and School Readiness programs (pp. 62-64, 81). The department should also identify ways to monitor the K-12 school performance of former participants in state-funded early childhood programs (p. 83).

Minnesota offers a range of early childhood programs, but state funding practices and oversight need further review.

Report Summary Minnesota funds and regulates a variety of child care and education-related programs that are targeted to prekindergarten children. We examined three large early childhood education programs— Head Start, Early Childhood Family Education (ECFE), and School Readiness. State funding for these three programs totals about $50 million for fiscal year 2001.

Programs Have Differing Designs, Costs, and Participant Characteristics

Head Start, ECFE, and School Readiness all aim to foster the healthy development of children. But, by design, these programs differ in their approaches. Head Start primarily serves low-income children ages three to five, while ECFE serves all ages of prekindergarten children and is open to all families without regard to income. Head Start provides many direct services to children, while ECFE relies more on parent education and service referrals. School Readiness is intended to fill the gaps not met by other programs, and it does this through a wide variety of activities directed toward children and parents. Children participating in Head Start are more likely to be poor, racial or ethnic minorities, and from one-parent families than children in ECFE or School Readiness.

On average, Head Start services are more comprehensive and intensive than ECFE and School Readiness services. Most Head Start programs provide 14 to 30 hours of direct services to children per week, for 9 to 12 months a year. In contrast, ECFE classes usually are two hours per week, and parents may attend classes for one to nine months a year. School Readiness services are determined partly on the basis of health and developmental screening, and nearly one-half of the children in School Readiness receive less than 30 hours of services during the year.

Service differences such as these explain most of the differences in the programs’ annual costs. 1 Statewide, Head Start costs over $5,000 per participating child, while ECFE and School Readiness cost less than $500 per child. Although Head Start’s costs per child are higher, its teacher salaries are one-half those paid to school districts’ ECFE and School Readiness teachers.

Minnesota’s Head Start Funding Is Generous but School Readiness Funding Is Modest

Minnesota is 1 of only 17 states that supplement the federal Head Start program with state funding. State funds account for about one-fourth of Minnesota’s total Head Start revenues, which is the fifth highest nationally. In 2000, state funds enabled Minnesota Head Start programs to serve nearly 3,000 more children than would otherwise have been served. The 1997 Legislature significantly increased Head Start funding, resulting in a corresponding increase in state-funded service hours.

Minnesota (with its School Readiness program) is 1 of 36 states with publicly funded preschool-type programs other than Head Start. Minnesota’s School Readiness spending per state resident under age five was below the spending median of the states with such programs. The state (Georgia) with the most generous funding for preschool programs other than Head Start spent four times more per state resident under age five than Minnesota spent for School Readiness and Head Start combined.

Early childhood programs use various approaches to foster healthy child development.

1 In addition, Head Start pays for some transportation and facilities costs that many ECFE and School Readiness programs do not bear.

State Head Start Funding Could Target Unmet Needs More Effectively

In fiscal year 2000, Head Start programs in Minnesota had funding sufficient to serve 45 percent of the state’s 1990 census estimate of children ages three to five in poverty. Only three states’ Head Start programs serve larger percentages of their children in poverty.

Minnesota law allocates one-half of the state’s Head Start appropriation on the basis of federal Head Start allocations, but the federal allocations do not adequately reflect the location of families in poverty. In 2000, service providers in the Twin Cities region were funded to serve about 33 percent of three-to five-year-olds in poverty, while providers in many rural areas of Minnesota were funded to serve more than 60 percent of the three-to five-year-olds in poverty.

Also, state Head Start funding allocations are not related to the types of programs that providers offer. Minnesota determines how many children each Head Start provider is authorized to serve by using federal cost-per-child rates that differ for each provider and are not based on present or proposed service costs. For instance, the number of children a provider is authorized to serve is not affected by whether the provider wishes to provide full-time or part-time services.

The Minnesota Legislature has authority to determine how state-appropriated Head Start funds are allocated, and it should consider an approach that relies less on federal allocation practices. The Legislature should also consider giving Head Start service providers more flexibility to determine which families to serve with their state funds— including families just above the federal Head Start income guidelines and families with children under age three.

Early Childhood Programs Are Widely Available, but Some School Districts Have Large Reserves

In the 1999-2000 school year, all but four Minnesota school districts offered ECFE services, and all but six offered School Readiness. About 24 percent of Minnesota’s children under age five participated in ongoing ECFE classes or home visits in 1998-99, and others were served through special family activities. Districts with School Readiness programs served 32 percent of Minnesota’s three-and four-year-olds. There is no clear, systematic basis for judging the extent to which the ECFE and School Readiness programs are serving Minnesota families interested in participating.

To ensure that districts use ECFE funds only for ECFE programs, state law requires districts to place all ECFE revenues in a reserve account. Many districts have accumulated large balances of unspent funds. The Minnesota Department of Children, Families, and Learning (CFL) has suggested that districts maintain 8 to 17 percent of annual ECFE revenues in reserve, but two-thirds of districts had more than 17 percent of their revenues in reserve in fiscal year 1999. Nearly a quarter of districts had more than 100 percent of annual ECFE revenues in reserve. Small districts with large reserves of unspent funds tended to provide fewer ECFE services than similarly-sized districts with small reserves.

SUMMARY xi Minnesota should distribute state Head Start funds in a manner that relies less on the federal allocation process.

Head Start Slots as a Percentage of Children Ages Three to Five in Poverty, By Service Provider

Over 60% 40 to 60% Under 40%

The Legislature should consider various options for addressing large reserve fund balances, including: (1) restricting ECFE funding for districts whose reserve fund balances exceed a certain level; (2) eliminating or reducing the statutorily guaranteed minimum funding levels for districts with under 150 eligible children; or (3) allocating a portion of ECFE funding based on the actual number of persons served, not just the size of the eligible population. In addition, there is a need for better CFL oversight of local ECFE and School Readiness programs— through improved financial monitoring and more on-site program reviews.

The Impact of Minnesota’s Programs Is Unclear

Nationally, various studies have shown that well-implemented early childhood programs can benefit children. However, the evidence is mixed about whether these benefits persist over time, and lessons from the research should be applied cautiously due to differences in programs from site to site. Studies have provided very limited evidence about the “right” age to intervene, the most effective types of services, and how long to intervene.

There have been many studies of Minnesota’s ECFE and School Readiness programs, but none provide definitive evidence about the impacts that can be attributed to these programs. Some individual districts systematically track the progress of preschool program graduates who have entered the K-12 school system, but this is not done on a statewide basis. The Legislature should ask CFL to report as soon as possible on steps that would be required to track the K-12 school success of children who once participated in state-funded early childhood programs.

Research has not definitively identified the “right” types of early childhood interventions.

Introduction In 1989, President Bush and state governors met to develop national education goals. The President and governors adopted a goal that “by the year 2000, all children will start school ready to learn.” 1 Their objectives toward this goal included the following:

All children will have access to high-quality and developmentally appropriate preschool programs that help prepare children for school.

Every parent in the United States will be a child’s first teacher and devote time each day to helping such parent’s preschool child learn, and parents will have access to the training and support parents need.

Minnesota has a variety of education-related programs for families with prekindergarten children. In April 2000, the Legislative Audit Commission directed our office to evaluate several of Minnesota’s early childhood education programs. In our study, we asked:

-What are the components of Minnesota’s early childhood education system? How are Minnesota’s programs similar and different, and how do these programs compare with those in other states? To what extent do Minnesota residents have access to these programs?

-How are Minnesota’s early childhood education programs funded? How has funding changed over time, and how have funds been spent? Should changes be made in the way state funds for these programs are allocated?

-What does previous research indicate about the effectiveness of early childhood education programs? What implications does research on child brain development have for state policy regarding these programs? Is there a need for further research on the outcomes of Minnesota’s programs?

-How adequately does the Minnesota Department of Children, Families, and Learning oversee the state’s early childhood education programs?

Legislators considered approving a study that would focus only on Minnesota’s Head Start program, but they decided the study should include other large, state-funded early childhood programs that have an education-related component. Our study focused on three programs: Head Start, Early Childhood Family

Our study focused on three programs.

1 The National Education Goals Panel, http://www.negp.gov/page1-7.htm; accessed August 29, 2000.

Education (ECFE), and School Readiness. We did not review child care subsidy programs or smaller early childhood programs.

The programs we examined are partly intended to help prepare prekindergarten children for later success in school. They do this by focusing on the broad goal of fostering the healthy physical, cognitive, social, and emotional development of children, and not just by preparing children academically for school. The programs encourage healthy child development in a variety of ways— through play and learning activities for children, parent education, home visits, family support services, child development and health screening, and other services. The term “early childhood education programs” may not fully reflect the broad scope of Head Start, ECFE, and School Readiness, but it is commonly used to describe programs of this type, and our report also uses this term.

To conduct this study, we relied on various sources of information. We obtained state and federal data on program finances, services, and participants from the Department of Children, Families, and Learning. We interviewed department staff and federal officials, and for many individual programs we reviewed plans, annual reports, monitoring reviews, and other documents on file with the department. We visited several early education programs around Minnesota to interview staff and observe program settings, and we interviewed many other program staff by phone. In addition, we collected information on parent fees from 41 school districts. 2 Finally, we reviewed national and Minnesota research studies related to child development and the outcomes of early childhood programs. Statewide data on the outcomes of Minnesota’s early childhood programs were not available.

Because there are a large number of early childhood programs in Minnesota, coordination is an important issue and we discussed this with some state and local officials. However, the Legislature directed the Department of Children, Families, and Learning to develop a plan for integrating child care and early childhood education programs, so we did not make this issue a major focus of our study. 3 The department is required to complete a report on program coordination by January 15, 2001.

Chapter 1 of this report provides a brief overview of Minnesota’s array of early childhood programs and discussions of Head Start, ECFE, and School Readiness. Chapter 2 discusses Head Start funding issues, including program availability and fund allocation methods. Chapter 3 discusses ECFE and School Readiness funding issues, including program availability, fund allocation methods, reserve funds, and state oversight of local financial practices. Chapter 4 summarizes previous research on the outcomes of early education programs. It also discusses the role of the Department of Children, Families, and Learning in ongoing program oversight.

Early childhood programs try to foster healthy physical, cognitive, social, and emotional development.

2 We requested ECFE and School Readiness fee information from a sample of 43 school districts, including the 10 largest districts and a random sample of 33 others.

3 Minn. Laws (1999), ch. 205, art. 1, sec. 61. The law says that “the plan must focus on costefficient delivery of services and address central location of programs, integration of programs, ease of accessibility to services by families, nontraditional hours of child care, infant care, sick child care, special needs child care, and legislative simplification of programs…. The plan must contain budget recommendations, proposed legislation in draft form, and recommendations for financial incentives to reward programs that provide cooperative services.”

1 Program Characteristics SUMMARY

Minnesota funds and regulates a variety of child care and education-related programs that are targeted to prekindergarten children. The three largest early childhood education programs— Head Start, Early Childhood Family Education (ECFE), and School Readiness— all aim to foster the healthy development of young children. By design, however, these three programs differ in their approaches. For instance, Head Start primarily serves low-income children, while ECFE is open to all families with prekindergarten children, regardless of income. Head Start provides many direct services to children and their families, while ECFE’s approach is more indirect— relying more on parent education and service referrals. Head Start has higher costs per participant than ECFE and School Readiness, reflecting its more intensive, comprehensive services and responsibility for facilities and other expenses that the other programs do not always bear. Head Start teacher salaries are about half those of teachers in ECFE and School Readiness.

Minnesota families make various arrangements for the care and education of their young children. Parents often rely on child care, preschools, or relatives to care for children while at work, and some parents choose not to work so they can stay home with their young children. Some families pay the full cost of their children’s care and education before kindergarten, while others rely in whole or part on public subsidies. 1

We focused our research on three of the largest state-funded programs that aim to foster healthy development and school readiness among prekindergarten children— Head Start, Early Childhood Family Education (ECFE), and School Readiness. In this chapter, we address the following questions:

-What is state government’s role in Minnesota’s system of services for prekindergarten children and their families?

-What are the characteristics of Minnesota’s three main early childhood education programs— Head Start, ECFE, and School Readiness? How do these programs compare in the services they provide, the types of people they serve, their costs per child, and their staffing arrangements?

1 There is statewide information on the number of children in various state-funded early childhood services, but there is not reliable statewide information on the extent to which parents use privatelyor locally-funded arrangements for their young children.

Throughout this chapter and the remainder of this report, we use the term “prekindergarten children” to refer to all children who have not yet started kindergarten— ranging from infants to children nearing kindergarten enrollment. Likewise, the term “prekindergarten programs” refers to all programs serving children in this age range.

OVERVIEW OF MINNESOTA’S EARLY CHILDHOOD SERVICE SYSTEM

In 1999, Minnesota had an estimated 322,000 children under age five, representing 6.7 percent of the state’s total population. 2 The Minnesota state demographer projects that by 2025 the number of children under five will decline to 282,000, or 5.4 percent of the total population. 3

The state plays two major roles in the care and education of preschool children. First, state licensing standards and laws regulate the way some early childhood services are provided. For instance, the state sets licensing standards for Minnesota’s 1,600 child care centers (including 500 public and private preschools) and 14,000 family day care facilities. 4 Child care centers have more stringent standards than family day care facilities, such as requirements that license-holders (1) hire teachers with specified education and experience levels; (2) limit the size of groups in which children’s activities occur; and (3) adopt program plans that outline strategies for promoting children’s physical, intellectual, social, and emotional development. 5 In addition, service providers must comply with requirements in state law to qualify for state funding. For example, the law specifies the amount that school districts must levy in local taxes if they wish to receive the maximum amount of state aid for their Early Childhood Family Education programs. 6

Second, the state provides financial support for the care and education of some preschool children. Table 1.1 shows the main programs that receive state funding. In programs such as Head Start, ECFE, and School Readiness, the state provides funds to organizations that provide services directly to children and families (or they contract with others to provide services). In child care programs, the state provides subsidies to eligible families, who then select the child care providers they wish to use. The programs in Table 1.1 that are operated on a statewide basis in Minnesota include:

-MFIP child care subsidies: Families enrolled in the state’s main welfare program— the Minnesota Family Investment Program (MFIP)— are eligible for child care subsidies. In addition, families remain eligible for up to 12 months of MFIP “transition year” child care after they stop receiving MFIP cash assistance. 7

-Basic Sliding Fee child care subsidies: Families who are not eligible for MFIP child care subsidies may be eligible for Basic Sliding Fee child care subsidies. 8 The state allocates Basic Sliding Fee funds to counties, which award subsidies to eligible families based on priorities set in state law. The amount of a family’s subsidy depends on its income. Up to 7 percent of the Basic Sliding Fee appropriations may be used for the At-Home Infant Child Care Program— which allows eligible families to receive a portion of their basic sliding fee subsidy while staying at home with an infant. 9

-Early Childhood Family Education: Local school districts receive state aid to operate programs open to all families with “children in the period of life from birth to kindergarten.” 10 The programs promote healthy child development, mainly through parent education and parent-child activities.

-Head Start: This program mainly serves low-income families. Using federal and state funds, Head Start promotes children’s cognitive, physical, emotional, and social development. It also helps participating families obtain the social services they need. Most of the enrolled children are ages three to five, but Head Start also serves some children under age three. 11

-School Readiness: Local school districts receive state aid for child development programs that promote success in school. Targeted toward “developmentally disadvantaged” and “at-risk” prekindergarten children over age 3½, School Readiness programs try to build on existing resources to meet children’s health, nutrition, education, and social services needs. 12

-Early childhood screening: State law requires that all school districts screen children before they enter school, targeting children between 3½ and 4 years old. At a minimum, the screening must include a developmental assessment, hearing and vision screening, immunization review, measurement of height and weight, identification of potential risk factors, an interview with a parent about the child, and appropriate referrals. 13

-Special education services: State law requires school districts to provide special instruction and services to preschool children with disabilities. 14

The state also has a role in various other early childhood programs that are provided at a limited number of locations. For instance:

-First Grade Preparedness: In the 2000-2001 school year, 33 school districts are receiving state aid to give children opportunities to develop skills before first grade that will help them succeed in school. 15 Eligible school sites (determined based on their percentage of children receiving free and reduced-price lunch) must offer full-day kindergarten for five-year-olds, programs for four-year-olds, or both.

-Way to Grow: In 2000-2001, the state is providing grants to public or private organizations in five cities to help pregnant women and families with children under age six meet the health and developmental needs of their children “at the earliest possible age.” 16

-Even Start: The state administers a site selection process for federal family literacy grants. In 2000-2001, programs at ten sites are offering adult education, parenting education, and early childhood education services for families with children under age eight. 17

-Family Service Collaborative Grants: The state awards grants for up to five years to local “collaboratives,” which must include a school district, county, public health agency, and community action agency. 18 These collaboratives design and implement integrated local service delivery systems for children and their families. There are collaboratives in 80 Minnesota counties, and about half of them receive state funds.

Nationally and in Minnesota, child care has often been funded separately from other early childhood programs, and some people have categorized individual early childhood programs as either providing “custodial care” or “education.” But, in practice, the differences among programs are often not as clear as such labels might suggest. Increasingly, various sorts of providers have implemented developmentally appropriate curricula and hired staff with child development training. In addition, some programs now combine traditional child care and supplemental services. For example, a Minneapolis program called Strong Beginnings subsidizes “regular” child care providers so they can add child development activities and family support services.

State government regulates and funds a wide array of early childhood services.

2 For July 1, 1999, see U.S. Census Bureau, http://www.census.gov/population/estimates/county/ca; accessed October 10, 2000. By comparison, 7.8 percent of Minnesota’s population was under five in 1990.

3 http://www.mnplan.state.mn.us/datanetweb/prj.html; accessed November 15, 2000. However, the annual number of births in Minnesota has been rising since 1995, contrary to the demographer’s projections.

4 State rules permit family day care facilities to serve up to ten children at a time, including up to six preschool children.

5 Minn. Rules, ch. 9503.

6 Minn. Stat. (2000) §124D. 135, subds. 3-4.

PROGRAM CHARACTERISTICS

Table 1.1: State and Federal Funding for Minnesota’s Early Childhood Programs, FY 2001

(Millions of $) Program State Federal

MFIP Child Care $78.61 $35.24 a Basic Sliding Fee Child Care 22.38 58.35 a Early Childhood Family Education 21.11 0.00 Head Start 18.75 64.16 School Readiness 10.40 0.00 First Grade Preparedness 7.00 0.00 Early Childhood Health and Developmental Screening 2.65 0.00 Family Services Collaboratives 2.44 0.00 Way to Grow 0.49 0.00 Early Childhood Special Education b b

a Includes state appropriations of federal Temporary Assistance for Needy Families transfers. Includes Minnesota Department of Children, Families, and Learning allocations of federal Child Care and Development Fund grants to these programs.

b Districts receive state and federal funding for special education, but services for many prekindergarten children are funded through grants to districts that are not distinct from those used to serve older children.

SOURCES: Minn. Laws( 1999), ch. 205, art. 1, secs. 71-72; ch. 241, art. 2, sec. 56, subd. 29; ch. 241, art. 2, sec. 60, subd. 25; Minn. Laws( 2000), ch. 489, art. 1, secs. 27, 30, 45; and Minnesota House of Representatives Research Department, Funding to Support Child Care Assistance( St. Paul, November 2000).

Minnesota’s early childhood programs receive more than $300 million per year in state and federal funds.

7 Minn. Stat. (2000) §119B. 05.

8 Minn. Stat. (2000) §119B. 03.

9 Minn. Stat. (2000) §119B. 061.

All Minnesota children receive developmental and health screening before entering kindergarten.

10 Minn. Stat. (2000) §124D. 13, subd. 1.

11 Minn. Stat. (2000) §§ 119A. 50-119A. 53.

12 Minn. Stat. (2000) §124D. 15.

13 Minn. Stat. (2000) §121A. 17.

14 Minn. Stat. (2000) §125A. 03.

15 Minn. Stat. (2000) §124D. 081. In addition, school districts can use a portion of their federal Title I funds to provide early childhood programs to low-income children, but the Department of Children, Families, and Learning estimates that only about 20 districts do so.

DESCRIPTION OF MINNESOTA’S THREE MAJOR PROGRAMS

In this section, we briefly describe each of the three early childhood education programs that are the focus of this study: Head Start, ECFE, and School Readiness. Later in this chapter, we directly compare these programs’ designs, participants, costs, and staffing arrangements.

Head Start

PROGRAM CHARACTERISTICS Child care and early childhood education programs generally have separate sources of funding, but their services are often similar.

16 Minn. Stat. (2000) §124D. 17. The cities are Minneapolis, St. Paul, Columbia Heights, St. Cloud, and Winona. Four of the five grantees are ECFE programs.

17 All current grantees are ECFE programs.

18 Minn. Stat. (2000) §124D. 23.

19 U.S. Department of Health and Human Services, Head Start Bureau, “Head Start Celebrating 35 Years, 1965-2000: The Head Start Story” (2000); http://www.hskids-tmsc.org/text/recruitment2/35thanniversary2/35history2.htm; accessed October 18, 2000.

In 1964, at the request of President Johnson’s administration, a panel of child development experts developed a blueprint to “help communities overcome the handicap of disadvantaged preschool children.” 19 Project Head Start, an eight-week summer demonstration program, was introduced the following year and became part of the government’s “war on poverty.” Today, Head Start operates as a 9 to 12 month program in all 50 states and several U.S. territories. In general,

-Head Start provides or arranges for a variety of education, health, and social services— primarily for families in poverty with children ages three to five.

Head Start programs are required to reserve 90 percent of their “slots” for children whose family income falls below the federal poverty line. 20 (Slots are the number of children that can be served at a given time by a Head Start program.) In addition, programs must reserve 10 percent of their slots for children with disabilities. Each Head Start program must develop criteria based on local assessments that define the types of children and families who will be given priority for services. Head Start children, once enrolled, must be allowed to remain in the program until they enter kindergarten.

Originally, the Head Start program exclusively served children ages three to five and their families. Today, there are state and federal competitive grant programs for Head Start services targeted at younger children, although Head Start continues to primarily serve children between the ages of three and five. 21

EARLY CHILDHOOD EDUCATION PROGRAMS At least 90 percent of Head Start participants must have family incomes below the federal poverty line.

20 The U.S. Department of Health and Human Services issues poverty guidelines each year. In 2000, a family of four met the poverty guidelines if their annual income fell below $17,050. For further information, see http://pr.aoa.dhhs.gov/network/FR-2000PovGuidelines.html.

21 In 1999, nearly 1,000 (7 percent) of the children served in Minnesota’s Head Start programs were under three years old.

Most of Minnesota’s Head Start programs are operated by community action agencies (CAA). CAAs are private, non-profit agencies that provide various services to low-income families. Compared with the country as a whole, a greater percentage of Minnesota’s Head Start programs are operated by CAAs and tribal governments, while a smaller percentage are operated by school districts or other public agencies. 22 In 1998-99, Head Start programs in Minnesota served about 15,000 children.

The federal government has established Head Start performance standards that outline the types of services programs must provide. For example, the standards require programs to individually assess children, implement a culturally-sensitive curriculum appropriate to the child’s developmental level, and make time for both adult-led and child-driven activities. Programs must provide activities such as art, music, physical education, and speech, and they must support emerging reading and math skills. 23 A typical Minnesota Head Start program provides children ages three to five with 14 or more hours of preschool and other services per week.

In addition to child development activities, the federal standards require programs to provide health, nutrition, and family services. For example, Head Start programs must ensure that children have had medical and dental screenings and received all recommended immunizations within 90 days of entering the program. 24 We found that:

-More than 90 percent of Minnesota’s Head Start children received medical screenings and all recommended immunizations in fiscal year 1999. However, only 72 percent received dental screenings, down from 82 percent in 1995.

Table 1.2 shows how Minnesota compared with Head Start programs nationally on selected service-related measures in fiscal year 1999. Minnesota’s rates of medical screenings and immunizations exceeded the national rates, but the state had a lower than average rate of dental screenings. Several Minnesota Head Start programs have expressed concern about the lack of dental providers willing to accept Medical Assistance reimbursements, the distance families must travel to dental providers, and strict policies regarding late or missed appointments that make it difficult for families to obtain services. 25

In addition to services to children, Head Start programs provide services and support to families. For example, Head Start programs identify the need for crisis intervention, mental health services, and family literacy services, and they work to secure those services for families. 26 Head Start programs also work with parents to assist with job training and employment; federal guidelines encourage Head Start programs to train and hire parents as Head Start staff. In Minnesota, nearly one-third of all Head Start staff are current or former parents of Head Start children.

A typical Head Start program provides at least 14 hours of preschool and other child services per week.

22 Of Minnesota’s 34 programs for three-to five-year-olds, 23 (68 percent) are operated by CAAs, 7 (21 percent) by tribal governments, 3 (9 percent) by non-profit agencies, and 1 (3 percent) by a school district. In the U.S. as a whole, 32 percent are operated by CAAs, 6 percent by tribal governments, 38 percent by non-profits, 17 percent by school districts, and 7 percent by other government agencies.

23 U.S. Department of Health and Human Services, Head Start Bureau, Head Start Program Performance Standards and Other Regulations (Washington, D. C., 1996), 60-71 (45 CFR 1304.21).

24 Ibid., 42 (45 CFR 1304.20).

25 Office of the Legislative Auditor’s analysis of the 1999 Department of Children, Families, and Learning survey of Head Start programs.

Head Start services are provided in several ways. Center-based services are provided in a classroom setting. Programs may choose to provide half-day or full-day services either four or five days per week. Families receive two home visits per year, typically conducted by the classroom teacher. Participating children receive at least 56 hours of service each month. In home-based services, program staff visit participating families once per week for at least 1½ hours and provide all Head Start components in the home. Families attend “group socialization” activities twice per month. Participants receive at least ten hours of service each month. Home-based services are more common in rural Minnesota counties than urban counties. Combination programs provide a blend of home and center-based services to families; programs must meet minimum service requirements of the home-based service option.

Table 1.2: Selected Measures of Head Start Services in Minnesota and U.S., FY 1999

Minnesota U.S. Percentage of children with:

Medical screenings 91 87 Dental screenings 72 84 All recommended immunizations 96 95 Medicaid enrollment a 67 62 More than one year of Head Start 33 25

Percentage of children who left after beginning the program 20 16 Percentage of open slots that were filled mid-year 82 84 Percentage of classes regularly assisted by volunteers 62 73

Number of slots per staff person 4.6 4.8 a Although both Head Start and Medicaid are federal poverty programs, many Head Start children are not enrolled in Medicaid because: (1) they have other insurance coverage, (2) their Medicaid coverage lapsed due to non-compliance with payment policies, or (3) they are non-citizens who are eligible for Head Start but ineligible for Medicaid.

SOURCE: Office of the Legislative Auditor’s analysis of federal Head Start Program Information Report data.

Minnesota trails the national average in Head Start dental screenings.

26 In 1999, Head Start implemented a family partnership agreement system which identifies family strengths, supports, and needs. In Minnesota, 85 percent of families have entered into a family partnership agreement with their Head Start program.

In addition to these options, programs may implement locally designed services if approved by the federal Head Start Bureau. Locally designed services vary in setting, duration, and method of service delivery; in one example, trained family day care providers offer Head Start services in their homes. 27 As shown in Table 1.3,

-Fifty-seven percent of Minnesota’s Head Start slots were provided in center-based settings in fiscal year 1999.

-Full-day, center-based services accounted for 14 percent of Minnesota’s Head Start slots in fiscal year 1999; grantees had more difficulty filling full-day slots than part-day slots.

Between 1995 and 1999, Head Start programs increasingly offered center-based services, while the number of home-based services declined somewhat. However, Minnesota still has proportionately more home-based services than do Head Start programs nationally; only 5 percent of Head Start slots nationwide are home-based, compared with 15 percent in Minnesota.

Early Childhood Family Education (ECFE)

Most states have targeted their prekindergarten programs to selected populations of children— typically, children from low-income families or children believed to be at risk of learning problems when they begin school. In contrast,

-Minnesota’s Early Childhood Family Education (ECFE) program is one of relatively few early childhood programs nationally that are open to all families, not just families with low incomes or at-risk children.

Table 1.3: Enrollment in Minnesota Head Start Programs, FY 1999

Percentage of Percentage of Funded Slots Slots Filled a Center-based slots 57% 104%

Full-week, full-day 14% 84% Full-week, part-day 2 104 Part-week, part-day 40 112

Home-based slots 15 111 Combination program slots 13 110 Locally designed option slots 15 96 TOTAL 100% 105%

a Percentages may exceed 100 if children leave the program mid-year and are replaced by other children or if a program over-enrolls for that program option.

SOURCE: Office of the Legislative Auditor’s analysis of federal Head Start Program Information Report data.

A majority of Head Start services are provided in center-based settings.

27 In 1999, 177 (1 percent) of Minnesota’s slots were family child care slots.

ECFE is a statewide program open to all Minnesota families with children who have not yet started kindergarten— ranging from infants to five-year-olds. Parents usually enroll in ECFE voluntarily, although some families involved in child protection or child custody cases are court-ordered to participate. As we discuss in Chapter 3, ECFE programs are available in nearly every school district. Districts provide ECFE services in a variety of locations— often in school buildings, but sometimes in homes, shopping centers, or other community sites.

ECFE aims to strengthen families by educating and supporting parents to help them provide the best possible environment for the healthy growth and development of their children. 28 While most nursery schools or day care programs focus primarily on direct services to children, the common denominator of ECFE programs throughout the state is what Minnesota law calls “parenting education” or “family education.” 29 By law, parents must be present in ECFE classes with their children or in concurrent classes while their children participate in ECFE activities at the service site. In general, services offered by ECFE are less comprehensive and less targeted than those offered by Head Start, as discussed further below.

An ECFE program typically consists of weekly two-hour classes. Individual classes usually last for less than 16 weeks, but some districts offer a sequence of classes that last throughout the school year. Each class is usually divided into two periods: one of parent-child interaction and one in which children and parents are separated. When the groups are separated, children participate in group activities while parents have a facilitated discussion about parenting or child development. Most districts offer several types of classes, often for specific age groups (see Table 1.4) or special populations (such as single parents or immigrant families).

In fiscal year 1999, 31 school districts collaborated with other districts to provide services or share administrative costs. Most of these were relatively small, rural school districts, but they also included some larger districts such as Bloomington and Richfield. As we discuss later in this chapter, districts often collaborate with other community programs to address the needs of families with young children.

Districts reported that nearly 78,500 children and over 87,000 parents participated in classes or home visits during the 1998-99 school year. Table 1.4 shows that classes for children of mixed ages were offered by most districts and had high parent attendance. Most programs provided some non-classroom activities, and programs reported that more than 60,000 children participated in ECFE activities besides classes and home visits in 1998-99. 30

EARLY CHILDHOOD EDUCATION PROGRAMS

ECFE emphasizes parent education and is open to all families with prekindergarten children.

A typical ECFE class meets for two hours per week.

28 Marsha R. Mueller, Immediate Outcomes of Lower-Income Participants in Minnesota’s Universal Access Early Childhood Family Education (St. Paul: Minnesota Department of Children, Families, and Learning, April 1996), 15.

29 Minn. Stat. (2000) §124D. 13, subd. 3.

30 Staff in several districts told us that providing an exact unduplicated count of the number of participants in special activities is difficult. For example, districts do not always require persons attending an open house to sign in.

School Readiness

School Readiness, originally called Learning Readiness, was established in 1991 by the Minnesota Legislature. School Readiness is designed to provide children “adequate opportunities to participate in child development programs that enable the children to enter school with the necessary skills and behavior and family stability and support to progress and flourish.” 31 As described in Chapter 3, nearly all school districts in Minnesota operate School Readiness programs. In 1998-99, districts reported that about 49,000 children participated in School Readiness programs.

The School Readiness program targets children between the ages of 3½ and 5. Districts may choose to serve younger children if they identify a local need for early intervention services. Programs are encouraged by law to make substantial outreach efforts to ensure that families with the greatest need receive services, including families with income levels below the federal poverty guidelines. Programs must give priority to children with developmental disadvantages or potential learning problems. 32 In addition, districts assess local service availability to identify children who are not being served by other programs, such as Head Start.

School Readiness gives priority to at-risk children.

Table 1.4: ECFE Services, FY 1999

Percentage Average Average of Programs Number of Parents Number of Children Offering Service Participating Participating Type of Service N= 311 Per Program Per Program

Parent-child classes

For children under 1 year 63% 48 44 For 1-year-olds 46 67 63 For 2-year-olds 47 71 66 For 3-year-olds 39 59 54 For 4-year-olds 40 59 61 For non-kindergarten 5-year-olds 25 63 66 For mixed-ages 82 181 208

Parent-only classes 42 136 -Activities Field trips 81 72 96 Open houses 75 78 78 Speakers 45 93 40 Parent-child activities 92 259 353 Home visits 59 54 71

NOTE: For activities, all participating children under age 18 were reported, but only children under five were reported for class participation data. Districts said that providing an exact, unduplicated count is difficult, particularly for non-class activities.

SOURCE: Office of the Legislative Auditor’s analysis of the Department of Children, Families, and Learning data from school districts’ annual ECFE reports.

31 Minn. Stat. (2000) §124D. 15, subd. 1.

32 Minn. Stat. (2000) §124D. 15, subd. 8. One way that districts identify children at risk is through a developmental screening, which children are required to receive within 90 days of entering the program.

State law requires School Readiness programs to address children’s social, cognitive, physical, emotional, and nutritional needs through referrals or direct services. Parents must be referred to adult literacy services, when needed. The law requires each program to have a comprehensive plan for meeting the needs of families with young children. 33 We found that:

-Half-day preschool is the most common School Readiness service, but the widely varying combination of services that districts offer in School Readiness makes it difficult to describe a typical program.

Table 1.5: School Readiness Services, FY 1999

Number of Programs Percentage Providing Service of Programs Service Description ( N= 321) Providing Service

Half-day preschool program 254 79% Collaboration with early childhood special education 244 76 Parent and child kindergarten preparation/transition 241 75 Life experience field trips 223 69 Transportation assistance 223 69 Contact with Head Start program 196 61 Take-home kits for children/parents 182 57 Consultants for children with specific needs in early childhood programs 143 45 Home visits with children and parents 142 44 One or more half-days of child-only time added to ECFE 126 39 Service for parents of children in existing programs 123 38 Collaboration with adult basic education/family literacy 122 38 Family resource center 110 34 Screening beyond basic early childhood screening 108 34 Special summer program 107 33 Supplementary services for children in preschools, child care centers, etc. 100 31 Programming with family child care providers 100 31 Story hour 77 24 Drop-in play times for children and parents 62 19 Interpreter assistance 53 17 Parent cooperative preschool 24 7 Other 101 31

SOURCE: Office of the Legislative Auditor’s analysis of data reported to the Department of Children, Families, and Learning by school districts in their School Readiness annual reports.

School districts use School Readiness funds for a wide variety of services.

33 Minn. Stat. (2000) §124D. 15, subd. 3.

Table 1.5 shows the percentage of School Readiness programs that offered various services in fiscal year 1999. Statewide, school districts reported offering an average of nine different services in 1998-99. The services ranged from daily preschool programs throughout the school year to summer programs to one-time field trips. State law encourages districts to provide transportation services, and over two-thirds offered some transportation assistance to participants. 34

State law encourages school districts to refer participating children to existing programs that can meet their needs. However, districts can establish new programs “where no existing, reasonably accessible program meets the program requirements.” 35 Each district’s School Readiness plan addresses how the district will collaborate with other agencies and early childhood programs to provide services.

DIFFERENCES AMONG HEAD START, ECFE, AND SCHOOL READINESS

The previous sections summarized the Head Start, ECFE, and School Readiness programs. Although all of these programs are often described as early childhood education programs, they are not interchangeable. Specifically, as we discuss in the next sections,

-Head Start, ECFE, and School Readiness vary considerably in program design, participant characteristics, cost per participant, and staffing arrangements.

Program Design

In some respects, Head Start, ECFE, and School Readiness all work toward a common purpose: to foster healthy social, emotional, and cognitive development of young children and prepare them to achieve in school. However, their approaches vary considerably. In particular,

-There are major differences in the design of Head Start and ECFE— with Head Start providing more intensive and comprehensive services to a more targeted population than ECFE.

Federal law requires that at least 90 percent of the children served by Head Start come from families with incomes below the federal poverty level. In contrast, ECFE is a “universal access” program that is intended for families in all income categories. A majority of children in Head Start participate in Head Start programs for at least 14 hours per week throughout the full school year. By comparison, a typical ECFE class meets for only two hours per week, and classes typically last for less than 16 weeks (although parents finishing one class may choose to enroll in another). Head Start provides or arranges for a wide array of services for children, including preschool classes, immunizations, health care, transportation, and nutritious meals. It also helps parents obtain job training, social services, and literacy training. ECFE, meanwhile, has a narrower scope— focusing largely on activities intended to educate parents and facilitate healthy parent-child interaction, and referring families to direct providers of other services.

Head Start, ECFE, and School Readiness overlap but are not interchangeable.

34 Minn. Stat. (2000) §124D. 15, subd. 4.

35 Minn. Stat. (2000) §124D. 15, subd. 5.

School Readiness fills some gaps between the approaches of Head Start and ECFE. Although School Readiness programs vary widely, most offer some preschool experiences to children, often for half-days, two or three times a week throughout the school year. In general, School Readiness programs address unmet needs by fostering collaboration among existing service providers or developing new services.

Minnesota’s early childhood education programs can also be distinguished by the extent to which they provide services directly to children (rather than indirectly, through their parents). ECFE is based on the notion that parents are children’s primary teachers, so it provides parents with skills and resources they can use in their daily interactions with their children. In contrast, Head Start and School Readiness provide more services directly to children.

Participant Characteristics

To compare participant characteristics among these three programs, we analyzed data collected by the Minnesota Department of Children, Families, and Learning and the U.S. Department of Health and Human Services. The three programs do not collect exactly the same information on participants, which hinders cross-program comparisons. In addition, the Head Start program collects demographic information on every Head Start participant (as required by federal law), while there is information on less than half of the ECFE and School Readiness participants (there is no state requirement for this information to be collected). We analyzed data on 30,000 ECFE families and 16,000 School Readiness families who returned participant questionnaires during the 1998-99 year, but we cannot be certain that the families in these samples are representative of all families participating in the program. Nevertheless, the data we examined suggest that:

-Children participating in Head Start are more likely to be poor, racial or ethnic minorities, and from one-parent families than are children in ECFE or School Readiness.

Table 1.6 summarizes participant characteristics. In general, the available data indicate that School Readiness participants more closely resemble participants in ECFE than those in Head Start. Some key areas of difference among these three programs include:

-Family income: Head Start serves families with much lower average incomes than either School Readiness or ECFE. Families participating in ECFE are somewhat more likely than families in School Readiness to have annual incomes above $50,000.

-Family structure: A large majority of the families participating in ECFE and School Readiness are two-parent families, while about half of the families participating in Head Start are two-parent families.

School Readiness fills some of the service gaps between Head Start and ECFE.

-Parent education: Participants in ECFE have slightly higher levels of educational attainment than those in School Readiness. This information is not collected for Head Start.

-Employment: Less than one-third of parents in Head Start and School Readiness are not working. While 40 percent of parents participating in ECFE are not working, nearly all of these parents reported that they are not seeking employment, suggesting that many are stay-at-home parents.

-Child’s age: Only 7 percent of Head Start participants during fiscal year 1999 were under age three, reflecting this program’s emphasis on services for children ages three and older. In contrast, over half of the children whose parents attended ECFE were under age three— presumably because the program is open to all parents with prekindergarten children, and its focus on parent education attracts parents of young children. 36

Table 1.6: Participant Characteristics of Early Childhood Programs, FY 1999

Head Start ECFE a School Readiness a

Percentage of families with: Annual income below $10,000 43% b 6% 8% Annual income above $20,000 --c 86 81 Annual income above $50,000 --43 30 Two parents 49 85 80

Percentage of parents:

With bachelor’s degree or higher --44 29 Not working d 29 40 28

Percentage of children who are:

Under 3 years old 7 53 e -Under 3½ years old -- --11 f Non-Hispanic white 50 87 83 Disabled 13 4 5

NOTE: Dashes indicate instances where data were not collected in participant questionnaires. a Data were collected from less than 40 percent of ECFE and School Readiness participants, and it is unclear whether the respondents were representative of all participants. b Forty-three percent of Head Start families reported incomes under $9,000. The Head Start questionnaire did not ask families whether their incomes were below $10,000. c Twenty-six percent of Head Start families reported incomes above $15,000. The Head Start questionnaire did not ask families whether their incomes were above $20,000. d Some “not working” parents are those who are not working by choice, such as stay-at-home parents. For ECFE, 88 percent of those who are not working are not working by choice; for School Readiness, 82 percent of these parents are not working by choice. Comparable data are not available for Head Start.

e Participating parents report the ages of all their children, including those in school. Among children under age six in ECFE families, we determined that 53 percent were under age three. f Because the School Readiness program is targeted to children over three-and-one-half, districts are asked to report the number of participating children under this age.

SOURCE: Office of the Legislative Auditor’s analysis of federal Head Start and Department of Children, Families, and Learning data.

Head Start serves more low-income families than ECFE and School Readiness.

-Race/ethnicity: Half of the children served by Head Start are racial or ethnic minorities. Both ECFE and School Readiness predominantly serve white, non-Hispanic participants.

-Disabilities: Head Start is required to reserve 10 percent of available slots for children that have been professionally diagnosed with a disability. In Minnesota, 13 percent of Head Start children have diagnosed disabilities. In contrast, 5 percent of the children served by School Readiness and 4 percent of the children served by ECFE are reported by school districts as having disabilities.

Cost Per Participant

Because Head Start, ECFE, and School Readiness are all early childhood education programs, legislators often compare the costs of these three programs. We found that:

-While cost per participant differs significantly among the Head Start, ECFE, and School Readiness programs, the differences largely reflect variation in the services provided.

To compare program costs, we examined available statewide data on state and local revenues for each program. 37 However, there are no reliable statewide data on the total amount of parent fees collected by districts for ECFE and School Readiness. As a result, the costs per participating child that we computed for ECFE and School Readiness somewhat understate the actual program costs— probably much more for School Readiness than for ECFE. 38

Largely reflecting differences in services provided, we found that statewide costs per child ranged from less than $500 for ECFE and School Readiness to just over $5,000 for Head Start. One reason for Head Start’s higher costs is that it provides a more comprehensive array of services to its participants. For instance, Head Start routinely provides child health screening and family social services; in contrast, ECFE and School Readiness are more likely to refer families to other agencies for these services rather than providing them directly.

Head Start’s longer classes and more comprehensive services help explain its higher costs.

36 We determined the percentage of children under age three from surveys of parents participating in ECFE. The surveys simply asked parents to specify the ages of their children, so the results likely included information on some children who did not directly participate in ECFE classes.

37 Head Start costs were calculated using the fiscal year 1999 state Head Start appropriation and total number of funded state slots; ECFE costs were calculated using fiscal year 1999 state aid and local levy revenues and the unduplicated number of children in classes or home visits; School Readiness costs were calculated using fiscal year 1999 state aid revenue and the number of total participants as reported in the 1999 annual report fee data.

38 According to budgets submitted by districts to the Department of Children, Families, and Learning, fees were expected to augment School Readiness state aid by 22 percent statewide in fiscal year 2001.

A second reason for cost differences is that Head Start services are typically longer and more intensive than ECFE and School Readiness services. Head Start provides a 9-to 12-month program for children, and its center-based program typically involves 14 to 30 hours of service per week. In contrast, ECFE classes are typically two hours per week, and parents may attend these classes from one to nine months during the year— although most individual classes are four months or less. School Readiness includes a wide variety of services, ranging from one-time consultations to half-day preschool services that last for the duration of the school year. Our discussions with program staff indicated that School Readiness’ overall average funding level of $241 per child is well below the average cost of providing year-long preschool services— but data collected by the state have not permitted precise determination of the cost per child for specific types of School Readiness services. Average funding levels per child in School Readiness vary considerably among school districts, likely depending on the types of services they offer. 39

A third reason for differences in cost per participant is that Head Start pays for some costs that ECFE and School Readiness typically do not. For instance, transportation costs accounted for up to 19 percent of the total costs of individual Minnesota Head Start providers in fiscal year 2000; in contrast, most ECFE programs have no transportation costs, and the availability of transportation services in School Readiness varies considerably from one district to the next. 40 Likewise, center-based Head Start programs are generally responsible for facilities costs, while school districts typically do not charge ECFE and School Readiness programs for the use of district facilities.

Overall, variation in services appears to explain most of the difference in cost per child between Head Start and the other two programs we examined. Unfortunately, data reported by school districts to the state have not permitted a comparison of the costs of providing specific service components among the three programs. The Minnesota Department of Children, Families, and Learning is compiling information on district-reported average costs for specific School Readiness services for 1999-2000, and this will be available in early 2001.

Head Start programs are more likely than ECFE and School Readiness programs to cover transportation and facility costs.

39 For example, Howard Lake-Waverly-Winsted reported an average funding level of $56 per child, largely because it counts every family that receives a newsletter as a participant. In contrast, Osseo, which offers preschool and transportation to participants, spent approximately $1,400 per child in fees and other revenues.

40 Three Rivers Community Action Agency reported that Head Start transportation comprised 18.9 percent of overall costs per child and 19.1 percent of center-based costs per child for fiscal year 2000.

Staffing Arrangements

We examined data collected by the Department of Children, Families, and Learning (CFL) and the federal government to analyze program staffing and salaries. Information was available for all Head Start staff, but data were available for only a minority of districts with ECFE and School Readiness programs. In addition, CFL does not distinguish between teachers in ECFE or School Readiness. The department has information about parent and child educators but these teachers may be teaching in ECFE, School Readiness, or both. 41 Because the ECFE and School Readiness data are incomplete and may not be representative of staff in all school districts, our comparisons with Head Start data should be considered tentative.

Some differences in staffing and salary patterns emerged in the available data. Specifically:

• Head Start teachers have lower educational attainment and lower average salaries than do ECFE and School Readiness teachers.

Table 1.7 compares staff characteristics in Head Start, ECFE, and School Readiness. It shows that when compared with Head Start, teachers in ECFE and School Readiness have higher educational attainment and much higher salaries. According to guidelines issued by the Department of Children, Families, and Learning, all ECFE teachers working with parents and children must be licensed teachers. 42 State law requires that School Readiness programs be supervised by a licensed or certified teacher, except for those services offered through contracts with private organizations. 43 State law does not require Head Start teachers to be licensed, but a new federal requirement states that 50 percent of Head Start teachers must have a child development or related degree by 2003. 44 In addition, Head Start programs give preference to current or former Head Start parents when hiring teachers.

Table 1.7: Staff Characteristics of Early Childhood Programs, FY 1999

Head Start ECFE/School Readiness Center-Based Parent Child Teachers Teachers Educators Educators

Average annual salary $14,740 $29,733 $34,787 Percentage with associate’s degree or higher 39% 100% 100% Percentage with bachelor’s degree or higher 33% 100% 100%

NOTE: Salaries for all staff are for full-time-equivalent positions. ECFE and School Readiness programs do not report educational attainment or salaries for all staff. Our analysis is based on salaries of 392 full-time-equivalent child educators reported by 139 districts and 143 full-timeequivalent parent educators reported by 110 districts. Percentages are based on staff for whom districts reported educational attainment. Our Head Start analysis is based on 535 teachers.

SOURCE: Office of the Legislative Auditor’s analysis of federal Head Start and Department of Children, Families, and Learning data.

On average, ECFE and School Readiness teacher salaries are twice those of Head Start teachers.

41 The number of districts reporting ECFE/School Readiness data totaled 139 for child educators, 110 for parent educators, 151 for ECFE coordinators, and 66 for School Readiness coordinators.

42 Department of Children, Families, and Learning, Teacher Licensure Clarification: Early Childhood Family Education (Roseville, MN, Fall 1999).

43 Minn. Stat. (2000) §124D. 15, subd. 10.

Coordinators of ECFE and School Readiness programs often work part-time or combine coordination and teaching duties. In contrast, Head Start directors work full-time, usually for the entire year, and are responsible for administering the entire Head Start program. Likely due to this variation in responsibilities and job intensity, Head Start directors average higher full-time-equivalent salaries and educational attainment than ECFE or School Readiness coordinators.

Summary of Program Differences

Table 1.8 highlights key differences in the Head Start, ECFE, and School Readiness programs. In particular, the table shows that the programs have different emphases, intensity, participants, and costs. Head Start typically provides the most intensive services, targeted toward a relatively small group of children and families in poverty. ECFE offers a more narrow set of services (primarily parent education) to a much broader population. The content of School Readiness services is more variable than ECFE and Head Start services from site to site because individual school districts design School Readiness programs to address gaps in local services.

SERVICE COORDINATION

As noted in the Introduction, our study did not focus on the coordination of Head Start, ECFE, School Readiness, and other early childhood programs because the Legislature asked the Department of Children, Families, and Learning to issue a report in January 2001 on coordination issues. 45 Nevertheless, we think it is important to state that:

-Minnesota, like many states, has a fragmented early childhood service system— with multiple programs and complex funding systems— and coordination of these services is an ongoing challenge.

In 1989, a report for the Governor’s Council on Children, Youth, and Families concluded that “the present system offers a jumbled array of overlapping and sometimes competing services whose overall quality is severely compromised.” 46 Likewise, a recent report by the Minnesota Early Care and Education Finance Commission concluded that “collaboration and communication between providers and programs need vast improvement.” 47

Head Start, ECFE, and School Readiness have different emphases, intensity, participants, and costs.

44 42 U.S. Code 9843a (a) (2) (A) (1998).

45 Minnesota Laws (1999) ch. 205, art. 1, sec. 61.

46 Task Force on Early Education and Care of Young Children, Putting It All Together: Building an Early Childhood System For Minnesota (St. Paul: Governor’s Council on Children, Youth, and Families, October 2, 1989), 4.

47 Early Care and Education Finance Commission, The Action Plan For Early Care and Education in Minnesota (Minneapolis, November 2000), 23.

 

Table 1.8: Program Differences Between Head Start, ECFE, and School Readiness

Head Start ECFE School Readiness Primary target population Children ages three to five from families in poverty

Families with children under age five Children ages three and four, with priority for “at risk” children

Number of children served in1998-99 15,000 78,500 (in classes or home visits)

49,000 Service provider Federal grantees that are non-profit or tribal agencies

School districts School districts Emphasis of service Provides comprehensive services to give disadvantaged children a “head start.”

Supports parents as the primary caregivers of their children.

Provides a range of activities to prepare children for school.

Primary services Preschool education, medical and dental referrals, immunization review, family crisis intervention and social services, and nutrition services.

Parent education, parent-child activities, referrals.

Varies from daily preschool to one-time workshops and referrals, depending on the district and participant.

Intensity of service Services last 9 to 12 months per year, for one to two years. Services range from 10 hours per month (home-based) to 56 to 120 hours per month (center based).

Classes typically last for less than 16 weeks, but participants may enroll in multiple classes throughout the year. Classes usually consist of weekly two-hour sessions.

Services range from one-time activities to nine months of preschool. Typically, preschool averages 60 hours per month.

Staff characteristics Teachers are usually unlicensed. About one-third are current or former Head Start parents. Average salaries are about $15,000 per year.

Teachers must be licensed. They average salaries over $30,000 per year.

Teachers are usually licensed. They average salaries over $30,000 per year.

Funding per child a $5,158 $ 474 $ 241 Total state funding,

FY 2001 (in millions) b $18.8 $21.1 $10.4 a Head Start funding includes administrative costs but excludes Birth to Three and Innovative Grants; ECFE funding includes state aid and local levy for children participating in classes or home visits. Does not include ECFE and School Readiness parent fees.

b ECFE funding includes state aid only. SOURCE: Office of the Legislative Auditor, 2000.

Unlike many states, Minnesota has all of its major early childhood programs under a single state administrative umbrella: the Department of Children, Families, and Learning. This is a potential advantage as the state examines ways to improve service coordination. But we also agree with a recent report that noted that “collaborative efforts are rooted at the community level, where prekindergarten programs operate and where families they serve live, work, and raise their children.” 48 In Minnesota,

-Local programs have several mechanisms for coordinating services, such as referrals and shared services, space, and staff.

For example, statewide surveys of Head Start agencies indicate that many Head Start programs collaborate with ECFE. Eleven of the state’s 34 Head Start programs reported that ECFE provided some portion of Head Start services in 1999, such as the group socialization aspect of home-based services. In addition, 14 Head Start programs reported that at least some of their services are located in the same building as ECFE. Half of the state’s Head Start programs reported that they refer all participants to ECFE. 49

State law requires School Readiness programs to work with existing service providers and build collaborative relationships. 50 We reviewed 37 School Readiness 1999 annual plan updates and found that 33 programs shared an advisory council or coordinator with ECFE, while only one shared an advisory council with Head Start. Nine of the 37 School Readiness programs reported that ECFE provided parent training for School Readiness participants.

Despite these local efforts to coordinate services, several providers told us that there is room for improvement in the level of collaboration among early childhood programs. For example, some staff told us about instances in which waiting lists have not been shared and transportation services have not been coordinated among service providers in the same geographic areas. More generally, some providers expressed concern that child care services are often funded separately from Minnesota’s other early childhood programs. We anticipate that the Department of Children, Families, and Learning will propose strategies to address problems such as these in its January 2001 service integration report to the Legislature.

Service coordination among Head Start, ECFE, School Readiness, and other early childhood service providers is an ongoing challenge.

48 Karen Schulman, Helen Blank, and Danielle Ewen, Seeds of Success: State Prekindergarten Initiatives, 1998-1999 (Washington, D. C.: Children’s Defense Fund, 1999), 157.

49 In addition, Head Start grantees usually work with other providers to secure medical, mental health, family literacy, and ancillary services for families because the grantees have limited resources to provide these directly.

50 Minn. Stat. (2000) §124D. 15, subd. 6.

2 Head Start Funding Issues SUMMARY

Minnesota serves a larger portion of poor families with its Head Start program than most states, but there is wide variation in the availability of Head Start services throughout the state. The Legislature should consider increasing the proportion of Minnesota’s state Head Start funding that is distributed based on a measure of unmet needs. Presently, half of state funds are allocated on the basis of federal Head Start allocations, and the federal allocations do not adequately reflect the costs of services provided or the location of families in poverty. In addition, the Legislature should consider giving Head Start grantees more flexibility to determine which families to serve with their state funds— including families just above the federal Head Start income guidelines and families with children under age three.

A s discussed in Chapter 1, Head Start is an early childhood education program targeted to children and families in poverty. This chapter further examines the Head Start program and discusses program availability, state funding, and recent expenditure trends. Specifically, it addresses the following questions:

• To what extent are Head Start services available throughout Minnesota, and is there evidence of unmet needs for services? To what extent does program enrollment vary around the state?

-How are state funds for Head Start allocated? What is known about the way grantees spend these funds?

-Did the recent increase in state Head Start funding result in a corresponding change in the amount of services provided?

To answer these questions, we used data from the federal Head Start Program Information Report and from the Department of Children, Families, and Learning (CFL). We also spoke with CFL staff and a number of Head Start service providers (commonly called “grantees”) across the state.

PROGRAM AVAILABILITY AND PARTICIPATION

Minnesota has 34 Head Start grantees that receive federal and state funds to serve children ages three to five. As shown in Figure 2.1, 27 of these grantees have federally-assigned, non-overlapping service areas, and together they provide services that cover Minnesota’s entire geographic area. In addition, seven American Indian tribal grantees and a grantee that serves migrant families provide Head Start services to children ages three to five in selected parts of the state. 1

Head Start grantees can serve children younger than three if they apply for and obtain federal or state funds specifically for children in that age category. In 1998-99, seven Minnesota Head Start grantees served children under age three with state funds, and four agencies enrolled children under three with federal funds. Statewide,

Figure 2.1: Service Areas of Minnesota’s Regular Head Start Grantees

NOTE: PICA (Parents in Community Action) is the Head Start provider in Hennepin County and SEMCAC (Southeastern Minnesota Community Action Council) is the Head Start provider in southeast Minnesota.

SOURCE: Department of Children, Families, and Learning. Tri-Valley

Head Start Northwest Head

Inter-County Head Start

West Central Minnesota Head Start Clay Wilkin Head Start Otter Tail-Wadena Head Start Western Head Start Southwestern

Prairie V Head Start

Tri-County Head Start Head Start Mahube

Reach-Up Head Start

Heartland Head Start Start Bi-County Head Start Kootasca Head Start Wright Cty. Head Start

Minnesota Valley Minnesota Head Start

PICA Scott-Carver-Three Rivers Head Start Dakota Head Start

Lakes and Pines Head Start

Anoka Ramsey

Action Programs

SEMCAC Head Start Arrowhead Head Start

Duluth Head Start Child Care Resource & Referral

There are federally designated Head Start agencies in all parts of Minnesota.

1 The grantee that provides services to migrant families (Tri-Valley Opportunity Council) is also 1 of the 27 grantees with a federally-designated service area.

-About 6.8 percent of Minnesota’s population of three-to five-year-olds were served in Head Start in 1998-99, compared with 0.5 percent of Minnesota’s population of children under age three.

To assess the availability of Head Start services for children ages three to five in Minnesota, we considered several sources of information. First, we used the Minnesota Head Start program’s definition of “unmet needs.” 2 Second, we looked at county-specific information on Head Start service locations, comparing each county’s number of Head Start “slots” with various measures of its population in poverty. 3 Third, we used data that grantees reported to the federal government to analyze the time it took to fill open Head Start slots in 1998-99. We also examined waiting list data for each grantee, although some state and grantee staff we spoke with questioned the reliability and consistency of this information as a measure of unmet need. Consequently, we do not report waiting list information here.

We found that: In program year 1999-2000, the total number of Head Start slots for three-to five-year-olds was equal to about 45 percent of Minnesota children ages three to five in poverty, as estimated by the 1990 census.

The Department of Children, Families, and Learning plans to use the 1990 census until 2002 or 2003 to determine Head Start fund allocations among grantees. However, there are reasons to be cautious about using data on the number of children in poverty in 1990 to compute unmet need. Head Start is a voluntary program, and program staff told us that some eligible families would choose not to participate even if the program was funded to serve all children in poverty. 4 Also, although 2000 census estimates are not yet available, other data from the U.S. Census Bureau suggest that, over the past ten years, there have been declines in Minnesota’s number of children under age five and overall poverty levels. 5 If so, relying on the 1990 estimates of child poverty may overstate the actual service needs. Using data on the actual number of children served by Head Start in 1998-99 and 1996 census estimates of Minnesota’s number of children in poverty, we estimated that Minnesota’s Head Start program now serves about 55 percent of children ages three to five in poverty. 6

HEAD START FUNDING ISSUES 27 Minnesota’s Head Start program serves about half of the eligible children in the state.

2 For state funding purposes, unmet needs are determined by a comparison of each grantee’s number of funded slots with the number of children in poverty ages three to five in the grantee’s service area. The number of children in poverty is based on data from the most recent decennial census.

3 Slots are the number of children a Head Start grantee can serve at any one time. We obtained information from the Department of Children, Families, and Learning about the location of Head Start service sites (and the number of slots at each) throughout Minnesota— including centers where Head Start classes are held and socialization sites where families enrolled in home-based services meet periodically.

4 Staff noted that some families choose not to participate because of cultural concerns about out-of-home care or because they prefer to participate in other early childhood programs.

5 For example, www.census.gov/hhes/poverty/povanin/pvmaptxt.html (trends in state and national poverty rates) and www.census.gov/population/estimates/state/st-99-09.txt (trends in state and national population, by age group).

6 Most children in Head Start are ages three and four at the start of the school year, and we estimated that Minnesota’s Head Start program served about three-fourths of three-and four-yearolds in poverty in 1998-99.

We found that: -Minnesota’s Head Start program serves a larger portion of children in poverty than most other states’ programs do.

Using data from several sources, we compared each state’s number of 1998 Head Start slots (funded by federal, state, and other sources) with its estimated number of children in poverty under age five. Only three states served larger percentages of children in poverty than Minnesota: Ohio and Delaware, which pay for Head Start slots with both federal and state funds, and Mississippi, which does not supplement its federal Head Start funds with state funds. 7

However, we found that: -Within Minnesota, the ratio of Head Start slots to poor children varies considerably.

For example, a grantee that serves a relatively low percentage of its estimated needs is the Anoka County Community Action Program, which serves Anoka and Washington counties. In 2000, it had 485 Head Start slots for three-to five-year-olds, which represented 31 percent of the service area’s estimated number of three-to five-year-olds in poverty in 1990. In fact, Minnesota’s four Head Start grantees that served the lowest proportion of estimated need were all in the Twin Cities metropolitan area, and all served less than 33 percent of the estimated number of three-to five-year-olds in poverty in 1990. 8 In contrast, Northwest Community Action Agency, which provides Head Start services in four counties in northwestern Minnesota, had slots that represented 116 percent of the service area’s estimated number of three-to five-year-olds in poverty. Six Head Start grantees had slots equaling more than 60 percent of their estimated need, and five of these grantees served counties on Minnesota’s western border. 9

We estimated local needs using two alternatives to the 1990 census: (1) census bureau estimates of the number of children under age 18 in poverty in 1995, and (2) the number of children receiving free and reduced-price lunch in Fall 1999. These alternatives have more recent measures of child poverty than the 1990 census, but they are reported only for broad age groups and not just for prekindergarten children. We compared these measures of need in each county with the county’s number of Head Start slots in fiscal year 2000; Table 2.1 shows the measures for selected counties, and Figure 2.2 shows the percentage of needs met by Head Start in each county (based on the 1990 census). 10 On both of the alternative measures, Head Start served lower percentages of needy children in each of the seven Twin Cities counties than in the state as a whole. 11 Statewide, the percentage of needy children served varied widely among individual counties, but the three measures in Table 2.1 showed similar patterns in the relative rankings of individual counties. 12 For instance, counties such as Benton and Wabasha had a small number of Head Start slots by any of the measures shown, while counties such as Renville and Wilkin had a large number of slots by any of the measures shown.

Head Start agencies in the Twin Cities area serve a relatively small portion of the region’s children in poverty.

7 We obtained data on federally-funded Head Start slots from the federal Head Start Bureau. For states that supplement federal Head Start funds with state funds, we used the total number of funded slots reported by Kathy Novak, Review of State Funding for Head Start (St. Paul: Minnesota House of Representatives Research Department, December 1999), 4. For the number of children in poverty, we used 1996 data (revised in November 2000) from the U.S. Census Bureau— http://www.census.gov/hhes/www/saipe/stcty/e96_00.htm; accessed December 7, 2000.

8 Besides Anoka County Community Action Program, the others were Parents in Community Action (serving Hennepin County), Ramsey Action Programs, and Scott-Carver-Dakota Community Action Programs.

9 Besides Northwest, the grantees over 60 percent included Prairie Five Community Action Council, Clay-Wilkin Opportunity Council, Wright County Community Action Council, Tri-Valley Opportunity Council, and West-Central Minnesota Communities Action Council.

 

HEAD START FUNDING ISSUES

Table 2.1: Head Start Slots By County, as a Percentage of Children in Poverty (Selected Counties)

Total 1999-2000 Head Start Slots, as a Percentage of: Head Total Free/Reduced-Estimated Estimated Start Price Lunches Persons under Persons under Slots, Provided per Age 18 in Age 5 in County 1999-2000 a Day, Fall 1999 b Poverty, 1995 Poverty, 1990

Benton 16 1.4% 1.6% 5.9% Wabasha 18 1.8 3.5 11.7 Red Lake 9 2.2 5.0 13.6 Rock 16 3.4 5.2 15.2 Dakota 295 3.6 5.3 17.1 Martin 48 3.6 6.1 14.3 Scott 63 3.7 6.2 26.6 Anoka 378 3.8 6.3 22.7 Meeker 64 4.0 9.3 25.5 Washington 133 4.1 5.0 17.0 Ramsey 1,538 4.2 7.4 19.0 Hennepin 2,109 4.4 5.9 18.0 Becker 196 11.4 11.7 34.0 Cook 19 13.0 17.4 39.6 Polk 287 13.1 17.6 53.2 Koochiching 96 13.9 15.2 50.8 Grant 56 14.1 23.7 83.6 Marshall 98 14.2 23.3 72.1 Norman 86 15.5 22.0 90.5 Wilkin 80 19.3 26.8 73.4 Renville 151 20.4 19.5 58.1 Mahnomen 211 21.7 51.2 150.7 STATE 13,067 5.9% 8.8% 26.6% a Slots were attributed to the county where the service center was located, although some of these centers provide service to two or more counties. b The analysis attributed lunches to the county in which the district office was located. There are a small number of instances in which the schools children attended were in a county different from the county where the district office was.

SOURCES: Office of the Legislative Auditor’s analysis of data from the Department of Children, Families, and Learning (number and location of Head Start slots, number and location of free and reduced-price lunches served) and U.S. Census Bureau (estimates of children in poverty).

10 To determine the number of Head Start slots in each county, we used information collected by the Department of Children, Families, and Learning that indicated the number of slots at each of the state’s Head Start service delivery sites.

11 Some of the 1995 county-specific census estimates have a sizable margin of error and should be considered tentative. In general, the census estimates are more reliable for larger counties than smaller ones, and the 1990 census estimates are more reliable than the 1995 estimates.

About half of Minnesota’s cities with population over 1,000 had a Head Start service site in 1999-2000, and there were also many cities and townships with populations under 1,000 that had service sites. As shown in Table 2.2, the largest city in outstate Minnesota without a Head Start site is Marshall, with an estimated 1999 population of 12,825. In contrast, 16 cities in the Twin Cities area with populations over 25,000 do not have Head Start sites. Not all cities need Head Start sites; some of the cities listed in Table 2.2 are relatively affluent, and low-income residents of some others may be able to obtain transportation to Head Start sites in nearby cities. For instance, although Brooklyn Center does not have a Head Start site within its boundaries, there is a Head Start center just outside the city limits (in Brooklyn Park). On the other hand, several cities listed in Table 2.2 had between 150 and 300 residents under age five in poverty in 1990, so there may be justification for additional Head Start locations. 13

 

Figure 2.2: Head Start Slots as a Percentage of Children Under Five in Poverty

SOURCE: Office of the Legislative Auditor’s analysis of 1990 census data and Department of Children, Families, and Learning data on 1999-2000 Head Start slots.

Over 60% 40 to 60% 20 to 40% Under 20%

The availability of Head Start services varies widely among Minnesota counties.

12 We correlated the alternative measures of Head Start needs with the measure based on 1990 census data. For the measure based on free and reduced-price lunch, r= 0.83. For the measure based on 1995 census estimates, r= 0.95.

Recently, Minnesota has had somewhat increased difficulty filling its funded Head Start slots in a timely manner, although openings tend to be filled more quickly in the Twin Cities region. Federal standards require grantees to fill vacant slots within 30 days. In the three most recent years for which data were available (1996-97 to 1998-99), 79 to 82 percent of vacant Head Start slots statewide were filled by grantees within 30 days, compared with 84 percent in the two previous years. The Head Start programs in Hennepin and Ramsey counties were two of the four programs in the state that filled all of their open slots within 30 days in 1998-99; the other grantees serving the seven-county metropolitan area also had above-average rates of filling vacant slots. 14 In contrast, most of the tribal grantees reported below-average rates of filling vacancies, as did five of the six grantees that were funded to serve more than 60 percent of their estimated needs.

HEAD START FUNDING ISSUES

Table 2.2: Largest Cities Without Head Start Service Sites, 1999-2000

In the Seven-County Twin Cities Area Outside the Seven-County Twin Cities Area 1999 Estimated 1999 Estimated City Population City Population

Bloomington 87,476 Marshall 12,825 Plymouth 64,313 St. Michael 8,291 Eagan 61,113 North Branch 6,772 Minnetonka 53,444 Waite Park 6,599 Eden Prairie 51,913 Baxter 5,475 Maple Grove 49,479 Stewartville 5,319 Edina 47,274 St. Joseph 4,934 St. Louis Park 44,236 La Crescent 4,778 Woodbury 42,342 Lake City 4,744 Maplewood 35,780 Kasson 4,370 Cottage Grove 31,137 Wadena 4,274 Fridley 28,623 Le Sueur 3,865 Brooklyn Center 28,597 Sleepy Eye 3,720 White Bear Lake 26,643 Goodview 3,415 Shoreview 26,545 Albertville 3,226 Oakdale 26,331 Byron 3,225

NOTE: Based on an analysis of service sites for children of all ages, including home-based socialization, center-based, and combination sites.

SOURCES: Office of the Legislative Auditor’s analysis of Head Start site data from the Department of Children, Families, and Learning and local population estimates from the State Demographer’s Office.

Outside the Twin Cities area, most cities with more than 3,000 residents have Head Start service sites.

13 Grantees have considerable latitude to plan the location and types of services they will provide within their service areas, although federal Head Start standards require grantees to conduct community assessments of their service areas every three years.

PROGRAM FUNDING Background

In all states, including Minnesota, federal Head Start funding goes directly to the grantees, bypassing state government. As a result, state legislatures have no authority to determine how a state’s federal Head Start allocation will be divided among individual grantees. 15 Each year, Head Start grantees receive an allocation from the federal government and are told how many slots they must provide for eligible children. The number of slots is determined by dividing each grantee’s total federal allocation by its current “cost-per-child” rate. When Head Start began in the 1960s, agencies submitted proposals to the federal government indicating how much it would cost them, per child, to provide Head Start services. These cost-per-child rates, adjusted over time for inflation and other factors, are still used to allocate funds today. 16 These rates vary by grantee, are largely based on historical factors, and do not reflect the actual costs of services presently provided.

The Minnesota Legislature first appropriated state funding for Head Start in 1988. Today,

-Minnesota is 1 of 17 states (plus the District of Columbia) that supplement the federal Head Start program with state funding. 17

In fiscal year 2000, state funds accounted for 24 percent of Minnesota’s total Head Start revenues, and they paid for nearly 3,000 slots. A recent report showed that four states funded a larger percentage of Head Start with state funds: Alaska (40 percent), Ohio (35 percent), Delaware (30 percent), and Oregon (30 percent). 18 Nine states (including Minnesota) use state funding to increase the number of children served, while the other states use their funds to meet federal requirements for local matching funds. 19

Federal Head Start funding goes directly to grantees and cannot be altered by state legislatures.

14 The other grantees with 100 percent “replacement rates” were Arrowhead Economic Opportunity Agency, which serves Cook, Lake, and St. Louis counties (excluding Duluth), and Otter Tail-Wadena Community Action Council. The Hennepin and Ramsey grantees had 100 percent rates in both 1997-98 and 1998-99.

15 As outlined in the federal Head Start Act, each state receives federal Head Start funding at least equal to the amount received in 1998. Funding beyond that level is distributed proportionately based on the number of children less than five years of age in poverty. At least 87 percent of the federal Head Start appropriation must be allocated according to this formula. Head Start Act, 42 U.S. Code, sec. 9801 (1998).

16 Individual grantees can request changes in their cost-per-child rates, but changes are relatively infrequent and there has been no system-wide review of these rates since the Head Start program was implemented.

17 Karen Schulman, Helen Blank, and Danielle Ewen, Seeds of Success: State Prekindergarten Initiatives, 1998-1999 (Washington, D. C.: Children’s Defense Fund, 1999) and Novak, Review of State Head Start Funding. In addition to Minnesota, the following states appropriate state funds for Head Start services: Alaska, Connecticut, Delaware, Hawaii, Kansas, Maine, Massachusetts, New Hampshire, New Jersey, New Mexico, Ohio, Oklahoma, Oregon, Rhode Island, Washington, and Wisconsin.

In fiscal year 1988, Minnesota provided $1.9 million for Head Start; in fiscal year 1990, state funding increased to $5.5 million. 20 Total state Head Start funding slowly increased until fiscal year 1998, when it jumped from $11.5 million to its current level of $18.75 million. 21 State Head Start funding has remained static at $18.75 million since fiscal year 1998. Table 2.3 shows the state and federal funding levels for Head Start since fiscal year 1988.

For the most part, Minnesota’s state Head Start funding is used to increase the number of Head Start slots each grantee provides; the only exceptions are state Head Start innovative grants. Minnesota statutes allow CFL to allocate up to 11 percent of state Head Start funds annually for innovative grants. 22 These grants are awarded on a competitive basis to (1) target Head Start resources to particular at-risk groups, or (2) provide services beyond those currently allowed under federal Head Start regulations. Since fiscal year 1991, innovative grants have comprised from 8 to 10 percent of total state Head Start funding. The percentage of innovative grant requests that were funded declined from 73 percent in fiscal year 1998 to 45 percent in fiscal year 2001, largely reflecting an increase in the amount of grant requests. 23 Table 2.4 lists the funding levels for innovative grants since 1991.

Table 2.3: Minnesota Head Start Funding, FY 1988-2001

State Funding Federal Funding Total Funding Fiscal Year (in Millions) (in Millions) (in Millions)

1988 $ 1.90 $12.60 $14.50 1989 2.40 14.80 17.20 1990 5.50 14.80 20.30 1991 5.50 17.60 23.10 1992 6.50 21.90 28.40 1993 8.50 25.00 33.50 1994 11.50 33.50 45.00 1995 11.50 40.50 52.00 1996 11.50 43.20 54.70 1997 11.50 43.70 55.20 1998 18.75 46.90 65.65 1999 18.75 53.65 72.40 2000 18.75 58.80 77.55 2001 18.75 64.16 82.91

NOTE: State funding includes all state Head Start funding (regular Head Start and Birth to Three programs, innovative grants, and state administration) and federal funding includes all federal Head Start funding (base funding, Early Head Start, expansion, etc.). State funding includes $1.5 million in 1988 and $1.4 million in 1989 of transfer funds from the Low Income Home Energy Assistance Block Grant.

SOURCE: Minnesota Department of Children, Families, and Learning, Head Start in Minnesota (Roseville, MN, January 2000), 2.

Compared with most states, Minnesota funds Head Start generously.

18 Novak, Review of State Head Start Funding, 4.

19 Ibid., 3. As per federal Head Start guidelines, grantees must provide a “local match” equal to 20 percent of the total cost of their program. Minnesota grantees generally use private or local funds or in-kind contributions to satisfy this requirement.

20 In fiscal year 1988, the Legislature directly allocated $400,000 to Head Start. An additional $1.5 million was transferred to Head Start from the Low Income Home Energy Assistance Block Grant.

21 Funding for Birth to Three, innovative grants, and state administration is included in the total state Head Start allocation.

22 Minn. Stat. (2000) §119A. 52 (b).

Most Minnesota Head Start grantees have received at least one innovative grant over the past five years. We found that:

-Between fiscal years 1997 and 2001, all but three Head Start grantees received at least one innovative grant.

Since 1997, grantees have received innovative grants for a wide variety of projects. Some grantees have used them for one-time purchases such as children’s books, playground equipment, or to facilitate collaboration with other organizations. On the other hand, five Head Start grantees have used the innovative grants to fund the same program every year between 1997 and 2001. 24 These programs include a mobile family service center to serve families in rural areas and a program that serves homeless children and their families.

Table 2.4: Head Start Innovative Grant Funding in Minnesota, FY 1991-2001

Total Innovative Grant Funding Innovative Grants as a Percentage Fiscal Year (in Millions) of Total State Head Start Funding

1991 $0.52 9.5% 1992 0.54 8.3 1993 0.71 8.3 1994 1.03 9.0 1995 1.10 9.6 1996 1.11 9.6 1997 1.07 9.3 1998 1.88 10.0 1999 1.71 9.1 2000 1.58 8.4 2001 1.61 8.6

SOURCE: Office of the Legislative Auditor’s analysis of Department of Children, Families, and Learning data.

Innovative grants comprise 8 to 10 percent of total state Head Start funding.

23 In fiscal year 1998, innovative grant requests totaled approximately $2.5 million; in fiscal year 2001, requests increased to over $3.5 million.

24 Anoka County Community Action Program, Arrowhead Economic Opportunity Agency, Parents in Community Action, Ramsey Action Programs, and Scott-Carver-Dakota Community Action Programs have all had ongoing programs funded through innovative grants since 1997.

25 Minn. Stat. (2000) §119A. 52 (a). Each grantee is guaranteed at least as much funding as it received in 1993, although this provision has not affected allocations since the 1998 funding increase.

Aside from innovative grants, Minnesota’s state Head Start appropriation is allocated to grantees based on a statutory formula. Minnesota allocates state Head Start funds among grantees based 50 percent on their share of the state’s federal funds and 50 percent on their portion of the state’s unmet needs. 25 These state funds are used to serve additional families in the Head Start program (beyond those served by federal funding). In our view, it is noteworthy that:

-The Legislature has authority to determine which organizations can receive state Head Start funds and how these funds should be allocated, but Minnesota’s approach is based largely on the federal Head Start funding allocation process.

Minnesota’s Head Start funding allocation process is closely tied to the federal process in the following ways: (1) only federal Head Start grantees are eligible to receive state Head Start funds; (2) each grantee’s number of state-funded slots depends on its federal per-child funding rate, which is used for federal allocation purposes; and (3) half of the state Head Start funds are allocated based on grantees’ shares of federal funding. In the following sections, we discuss each of these aspects of Minnesota’s Head Start funding formula, as well as policy alternatives that the Legislature could consider.

Eligible Head Start Grantees

According to Minnesota law, only grantees that existed in 1989 are eligible to receive state Head Start funding. 26 As a result, all Minnesota state Head Start grantees are also federal Head Start grantees. Four states have taken a different approach by allocating state Head Start funds to federal grantees as well as other organizations. 27 Through competitive grants, Delaware allocates some of its state Head Start funds to public and private preschool programs in priority regions of the state. This is part of the state’s initiative to provide children in families with below-poverty incomes one year of preschool and reduce the waiting lists at Head Start centers. New Jersey provides funds to community child care programs to help them meet some of the Head Start standards. Ohio and Oregon provide funding to non-federal Head Start grantees that agree to follow the Head Start standards.

The Legislature may wish to consider allowing agencies other than federal Head Start grantees to be eligible for state-funded innovative grants. Historically, innovative grants have not been awarded to provide additional Head Start sites. However, as discussed earlier in this chapter, there are many large cities in Minnesota that do not have a Head Start program, and there are some areas of the state where Head Start programs serve a relatively small portion of the eligible children. If existing grantees do not provide services in these areas, there might be an opportunity for other early childhood educators or care providers to offer a Head Start program.

RECOMMENDATION

The Legislature should consider amending Minn. Stat. §119A. 52 to authorize agencies other than those that received Head Start funding in 1989 to be eligible for innovative grants.

Minnesota law permits only federal Head Start grantees to receive state Head Start funding.

26 Minn. Stat. (2000) §119A. 52 (a).

27 Schulman and others, Seeds of Success, 49, 184, 200, and 202.

If the Legislature does this, it should also consider whether the non-federal grantees should be required to meet all federal Head Start standards in the programs they offer. Some officials from Minnesota Head Start programs told us that they could provide high-quality services to more eligible recipients at lower costs if their state-funded services were not required to meet all federal requirements. Such flexibility might be consistent with the law’s stated purpose for state Head Start innovative grants, which includes providing “services in addition to those currently allowable under federal Head Start regulations.” 28 On the other hand, the comprehensive federal requirements have traditionally defined “true” Head Start services, so the state should be cautious before it allows a diluted version of an established program.

Per-Child Funding Rates

Every biennium, Minnesota Head Start grantees receive a funding allocation from the Department of Children, Families, and Learning. Based on state law, the department divides each grantee’s state allocation by its federal per-child funding rate to determine the number of slots each grantee must provide. In fiscal year 2000, Minnesota’s per-child rates varied from about $4,500 to $7,000 among grantees. 29 As discussed earlier, these cost-per-child figures are largely a vestige of the original federal Head Start funding applications from the 1960s and do not necessarily reflect present differences in grantees’ actual costs or services provided. In fact, we found that:

-In Minnesota, a grantee’s state funding level and number of slots are not related to the type of program a grantee offers.

For example, if a grantee receives $50,000 in state Head Start funding and has a federal cost per child of $5,000, the grantee will be expected to provide ten state-funded slots, regardless of the type of programming offered. Under the current funding structure, the grantee could provide ten full-day, full-week slots or ten half-day, part-week slots and receive the same level of state Head Start revenue.

Minnesota allocates state Head Start funds based on federal per-child funding rates.

28 Minn. Stat. (2000) §119A. 52 (b).

29 One tribal grantee has a per-child rate of over $10,000 but it had only 19 state and federal slots in 2000.

Not surprisingly, the cost differences among programs can be significant. Based on our review of detailed expenditure data from six Head Start grantees, it is clear that costs per child vary both by provider and type of service provided. 30 Among these six grantees, actual average costs per child ranged from $5,900 to over $7,710 in fiscal year 1999. When comparing costs among types of service provided, center-based services were more costly than home-based or combination programs. For one grantee, center-based services cost $8,075 per child, while services combining a center-based and home-based approach cost only $3,700 per child.

Three states (Connecticut, Ohio, and Wisconsin) use a distribution method other than the federal per-child amount to allocate state Head Start funds. Connecticut allocates funding to provide additional classrooms, rather than funding programs on a per-slot basis. Ohio provides a per-child amount that is less than the federal amount, based on a legislative formula. Wisconsin provides a per-child amount based on the statewide average of its grantees’ federal per-child rates. 31 We think the Legislature should consider an alternative approach to Minnesota’s present method of allocating state Head Start funds.

RECOMMENDATION

The Legislature should consider amending Minn. Stat. §119A. 52 to authorize the Department of Children, Families, and Learning to determine a grantee’s number of state-funded slots based on its proposed services and the cost per child of those services, rather than based on the federally-set rates per child.

If such a change were made, state funding would be more reflective of the services and programs offered by each grantee. This might encourage grantees to provide longer hours of service or, in some cases, serve more families. However, such a change would require CFL to monitor programs more closely. Specifically, CFL staff would have to ensure that the programs grantees proposed are the programs ultimately offered. 32

State Head Start Funding Formula

State Head Start funds are distributed among grantees based 50 percent on their share of the state’s federal funds and 50 percent on their portion of the state’s unmet needs. State officials told us that half of the allocation was based on each grantee’s share of federal funding to give grantees an incentive to maximize their federal funds. However, federal funding has been fairly inflexible over time, largely reflecting historically-negotiated rates rather than present costs and service needs. Grantees can apply to the federal government for changes in funding rates or number of slots, but people we spoke with said that these changes have had fairly limited effects in recent years on grantee funding levels. 33

Variation in the actual cost of providing services is not considered in the allocation of Minnesota’s state Head Start funds.

30 We obtained detailed cost data for Clay-Wilkin Opportunity Council, Heartland Community Action Agency, Lakes and Pines Community Action Council, Mahube Community Council, Northwest Community Action, and Three Rivers Community Action.

31 Novak, Review of State Funding for Head Start, 3.

32 In addition, programs may have less flexibility in changing the type of services they offer once a budget has been approved.

The other portion of state Head Start funding, which is based on the unmet need in each grantee’s service area, helps target state funds to areas most in need of Head Start services. To calculate the unmet need for Head Start funding, the Department of Children, Families, and Learning uses the most recent census data to determine the number of children ages three to five in poverty in each service area. CFL computes unmet need by comparing this number with the number of children being served with federal Head Start dollars. As discussed earlier, there is wide variation in the extent to which these needs are funded throughout the state. In 2000, the percentage of eligible children served with both federal and state Head Start funds ranged from 31 to 116 percent of the estimated need in grantee service areas.

RECOMMENDATION

The Legislature should consider increasing the percentage of state Head Start funds allocated on the basis of unmet needs.

Such a change would help to improve the equity of Head Start allocations across the state by targeting funds to areas with higher levels of need. However, it is worth considering that this change would likely increase the percentage of state funding going to grantees in the Twin Cities area. Some Head Start officials told us that there are many services and programs targeted toward low-income families in the metropolitan area, whereas outstate Head Start agencies may provide the only services for low-income families in their communities.

If the Legislature adopts a more need-based Head Start allocation formula, it may wish to consider more up-to-date measures of need than the decennial census. The department expects to use the 1990 census as a basis for state Head Start allocations through 2002 or 2003. However, it is not unusual for the state to experience significant population shifts and changes in regional economic conditions between censuses. 34 The Legislature could consider an alternative measure of child poverty, such as the number of free and reduced-price lunches (a measure which is calculated annually and currently used in the allocation formula for the School Readiness program).

Minnesota could more effectively target Head Start funds, based on measures of unmet needs.

33 Some Minnesota grantees were successful in obtaining additional funding from the federal government in the early 1990s. However, there has been little new Head Start money available from the federal government since that time.

34 Between the 1980 and 1990 censuses, the number of three-to five-year-olds in poverty in Minnesota increased 33 percent. In individual Head Start grantees’ service areas, however, the 1980-90 change ranged from a 33 percent decline to a 94 percent increase.

IMPACT OF RECENT FUNDING INCREASES

In fiscal year 1997, the state appropriated $11.5 million in Head Start funding; in 1998, the state allocated $18.75 million for Head Start, a 61 percent increase once adjusted for inflation. 35 Since 1998, the state has allocated $18.75 million each year for Head Start. Head Start grantees report expenditures of state funds to the Department of Children, Families, and Learning using 11 cost categories including personnel, fringe benefits, travel, supplies, and construction. We examined these data and found that:

• In general, Head Start grantees spent their state funding in roughly the same proportions over the past four years.

Specifically, expenditures on personnel and fringe benefits represented approximately 76 percent of the budget in both fiscal years 1997 and 2000. Similarly, travel expenditures comprised over 2 percent of the budget and supplies accounted for about 5 percent in both 1997 and 2000. Some cost categories, particularly equipment and construction, did fluctuate during this four-year time period. However, these categories are more likely than others to contain varying or one-time expenses. For example, equipment costs include expenditures on buses and computer networking.

State Head Start funding increased significantly in 1998.

35 The $18.75 million in state appropriations includes funding for regular and Birth to Three Head Start programs, innovative grants, and state administration. Adjusted for inflation using the CPI-W. Bureau of Labor Statistics, Consumer Price Index – Urban Wage Earners and Clerical Workers (2000); http://stats.bls.gov/top20.html; accessed September 4, 2000.

Because spending across expense categories increased in proportion to previous spending levels, it appears that grantees used the state funding increase to provide more services to Head Start children and families. To examine this further, we reviewed changes in slots and “contact hours” between program years 1997 and 1999. Contact hours measure the hours of Head Start service provided to children. We based our estimates on federal requirements for minimum hours of service. These requirements vary among different types of Head Start program options. For example, center-based, full-week, full-day programs are required to provide 963 service hours per year while home-based programs are required to provide only 72 service hours per year. 36 When all funding is considered (from federal, state, private, and local sources), we found that:

-Between 1997 and 1999, the total number of slots and contact hours provided by Minnesota Head Start grantees increased.

As shown in Table 2.5, the total number of funded slots increased by 8 percent and the total number of contact hours increased by 17 percent. 37 This means that Minnesota grantees served more families and provided more intensive services in 1999 than in 1997. However, as we discuss in the remainder of this section, between 1997 and 1999, state-funded slots and contact hours increased considerably more than those funded by the federal government. In addition, funding from local and private sources decreased during this same time period.

We found that: State Head Start funding in fiscal years 1998 and 1999 was used both to increase the number of families served and to increase the hours of service provided.

Between fiscal years 1997 and 1999, regular state-funded Head Start contact hours increased by 51 percent and state-funded slots increased by 38 percent. Because contact hours increased more than slots, it suggests that Head Start grantees, on average, are providing more intensive services to Head Start families than they had previously. According to our analysis, the largest percentage increase in funded contact hours between 1997 and 1999 came through an increase in full-week, part-day programs. 38

Table 2.5: Changes in Minnesota Head Start Funded Slots and Contact Hours, FY 1997-99

Percentage Change in Funded: Funding Source Slots Contact Hours

Federal 4% 13% State 38 51 Local and Private -45 -43

TOTAL 8% 17% NOTE: Excludes migrant and Early Head Start data. SOURCE: Office of the Legislative Auditor’s analysis of federal Head Start Program Information Report data.

Minnesota’s Head Start services expanded in recent years— mainly due to the increase in state funding.

36 U.S. Department of Health and Human Services, Head Start Bureau, Head Start Program Performance Standards and Other Regulations (Washington, D. C., 1996), 249-254 (45 CFR 1306.32 – 1306.34).

37 These data exclude migrant and Early Head Start slots and contact hours.

38 Using our estimates of funded contact hours based on the federal minimum requirements, between 1997 and 1999 center-based, full-week, part-day contact hours increased by almost 200 percent and center-based, full-week, full-day contact hours increased by almost 150 percent. Contact hours for all other program options increased by less than 50 percent during this same time period.

We also found that: Between fiscal years 1997 and 1999, state-funded Head Start contact hours increased at a rate comparable to the increase in state Head Start funding.

Between fiscal years 1997 and 1999, state Head Start revenues for regular Head Start programs (excluding funding for innovative grants, Birth to Three programs, and state administration) increased by 49 percent, once adjusted for inflation, while contact hours increased by 51 percent during the same time period. The Minnesota Legislature increased state Head Start funding significantly in fiscal year 1998 to increase services provided. In our view, the Legislature should be reassured that the funding increase led to a comparable increase in estimated state-funded Head Start contact hours.

Similarly, the increase in federal-funded contact hours between 1997 and 1999 was comparable to the increase in federal funding for the same time period. Regular federal Head Start funding increased by about 11 percent between 1997 and 1999, once adjusted for inflation, while federal-funded contact hours increased 13 percent. Although both federal-and state-funded Head Start slots and contact hours increased over the past few years, we found that:

-Slots and contact hours funded by private and local sources decreased significantly between 1997 and 1999.

During this time period, the number of slots funded by private and local sources (such as local governments, foundations, and grants) decreased 45 percent and contact hours decreased 43 percent. 39 As a result, some of the recent increase in state-funded Head Start slots was offset by a decrease in slots paid for with funds other than state and federal revenues. Grantees told us that they try to maximize their funding from all sources, and we have no evidence that the decrease in private and local funding resulted from the increase in state Head Start funds.

Finally, we found that: -Between 1997 and 1999, federal and state funded Head Start slots and contact hours increased more than filled slots and contact hours.

This is noteworthy and may imply that grantees are having difficulty filling all of their existing slots. Throughout our evaluation, CFL staff and grantees commented on problems with fully enrolling Head Start programs. Some Head Start grantees indicated that although they are unable to fill all of their Head Start slots for three-to five-year-olds, they have large waiting lists for programs serving children under age three. Federal and state funding is currently given to grantees in two separate allocations – one for “regular” Head Start services provided by all grantees for children ages three to five, and one for services that selected grantees provide for children under age three. 40 If state Head Start funding could be used for all eligible children under age six, it would give grantees increased flexibility to assist those families most needing services.

Allocating Head Start funds by age group limits grantees’ flexibility.

39 One grantee accounted for almost 40 percent of the total decrease in slots funded by private and local sources.

Many grantees also told us that there are an increasing number of families interested in Head Start who earn between 100 and 125 percent of the federal poverty level and are therefore not eligible for the program. 41 Grantees and CFL staff attribute this recent phenomenon to welfare reform and the increased number of families earning income. The Minnesota Legislature could broaden the income-eligibility guidelines for state Head Start funds so that more families just above the poverty level could be served. This would provide access to more families who could benefit from Head Start services and help grantees to achieve full enrollment. Currently, Ohio is the only state that allows state Head Start funds to be used to serve families with incomes up to 125 percent of poverty. 42

RECOMMENDATION

The Legislature should consider: (1) allocating state Head Start funding to be used for all eligible children under age six, allowing grantees to determine how to allocate their funding among children of various ages, and (2) broadening the income eligibility guidelines for state-funded Head Start slots.

MONITORING OF HEAD START FINANCES

At the outset of our study, some legislators questioned whether there is sufficient financial oversight of individual Head Start programs. In our judgment,

-Federal and state officials generally provide adequate oversight of local Head Start programs’ finances.

Increasingly, families interested in Head Start are just over the federal Head Start income guidelines.

40 Early Head Start is the federal program for children ages birth to three; Birth to Three is the state program for serving children in this age group. Grantees are required to apply for these funds separately from their regular Head Start allocations. State Birth to Three funds are competitive two-year grants; Early Head Start funds were initially awarded as five-year projects, but have since become part of grantees’ annual base allocation.

41 In the 1999 Head Start Questionnaire collected by CFL, grantees indicated that they received an average of 69 over-income applicants, most of whom were within 150 percent of the income limits.

42 Schulman and others, Seeds of Success, 201.

Federal and state officials, in accordance with Head Start regulations and statutes, regularly collect and monitor information on Head Start grantees’ programs. Head Start grantees are required to report their state expenditures at least quarterly to CFL and are not allowed to carry over unspent state funds into the next biennium. Grantees must itemize and report their state expenditures by cost categories including salaries, benefits, travel, equipment, and construction. Head Start expenditures are reviewed as part of the regular federal and state program review process and must meet federal Head Start standards. In addition, Head Start programs are audited annually by an independent auditor to determine if the agency’s financial statements are accurate, if the agency is complying with the terms and conditions of the grant, and if the agency’s financial and administrative procedures are effective. 43

State officials could do more comparisons of individual Head Start grantees’ program costs. For example, CFL could examine the cost differences among grantees for similar services. However, overall we think the department has an adequate foundation of financial information on Head Start programs.

43 Head Start Program Performance Standards, 6 (45 CFR 1301.12).

3 ECFE and School Readiness Funding Issues

SUMMARY

Statewide, Early Childhood Family Education (ECFE) programs served 24 percent of children under age five in 1998-99 in weekly classes or home visits, and School Readiness programs served 32 percent of children ages three and four. The participation rates of individual districts varied widely, and there is no reliable statewide information on the extent of unmet needs for these programs. In 1999, two-thirds of districts had ECFE reserve fund balances that exceeded guidelines subsequently established by the Department of Children, Families, and Learning. The Legislature should consider the merits of various options for changing ECFE funding practices, including (1) restricting funding for districts with large ECFE fund balances; (2) reducing guaranteed minimum funding levels; and (3) allocating a portion of ECFE funding based on the actual number of persons served, not just the size of the eligible population. The Legislature should also consider establishing a reserve fund for School Readiness, and the department should monitor district ECFE and School Readiness finances more closely.

As discussed in Chapter 1, Early Childhood Family Education (ECFE) and School Readiness are administered by local Minnesota school districts. This chapter examines these two programs in further detail, and it addresses the following questions:

• To what extent are ECFE and School Readiness programs available throughout Minnesota, and is there evidence of unmet needs for services? To what extent does program enrollment vary around the state?

-How are state funds for ECFE and School Readiness allocated?

-What is known about the way districts spend ECFE and School Readiness funds? To what extent do districts have reserves of unspent funds?

To answer these questions, we reviewed data collected by the Department of Children, Families, and Learning (CFL), including annual ECFE and School Readiness reports, Uniform Financial Accounting and Reporting Standards (UFARS) data, and revenue data. We also interviewed staff at CFL and in school districts across the state, and we conducted a survey of 41 ECFE and School Readiness programs regarding their fee structures.

ECFE PROGRAM AVAILABILITY AND PARTICIPATION

ECFE has grown considerably since its start as a pilot program in six school districts in 1974. We found that:

-All but 4 of Minnesota’s 347 school districts operated ECFE programs in 1999-2000.

The only districts that did not have an ECFE program were Walnut Grove, Butterfield, Franconia, and Prinsburg. 1 Together, these districts reported only 122 children under age five living within their boundaries in 1998-99. Some school districts operate programs jointly with other districts, but officials from the Department of Children, Families, and Learning told us that each district that participates in the program has at least one site within its boundaries where ECFE services are provided.

To evaluate the level of participation in ECFE programs, we analyzed data that school districts report annually to CFL regarding the number of (1) ECFE participants, and (2) children in the district under age five. Specifically, we analyzed the reported unduplicated number of persons participating in ECFE classes or home visits during the 1998-99 year, and we excluded from our analysis persons reported as participating only in ECFE special events or activities. 2 We found that:

-Statewide, about 24 percent of children under age five participated in ECFE classes or home visits during 1998-99.

-Individual districts varied considerably in their reported participation levels— from less than 10 percent of children under age five to more than 50 percent.

Table 3.1 shows districts with more than 500 residents under age five that served the highest and lowest percentages of children in ECFE during 1998-99. Staff in these districts provided a variety of explanations for their participation levels. Fergus Falls ECFE staff attributed the district’s high participation rate to its relatively short classes (some are only four weeks long) and its use of non-governmental grants to expand program offerings. Columbia Heights ECFE staff said that the district’s high reported level of participation in 1998-99 was probably inaccurate, and more careful counts in 1999-2000 resulted in a 60 percent decline in reported participation. In the Pipestone-Jasper district, ECFE-sponsored visits to the homes of all families with newborn children may partly explain high participation rates. Staff in St. Francis attributed the school district’s low rate of ECFE participation in classes and home visits partly to an inability to find space for expanded course offerings. Staff in St. Cloud cited difficulties finding qualified teachers willing to work Saturdays and evenings, as well as difficulty finding culturally diverse staff to serve the community’s increasing number of minority families.

ECFE services are widely available throughout Minnesota.

1 Franconia and Prinsburg are classified as “non-operating” districts – which means they do not operate public schools.

2 Districts are supposed to annually report the unduplicated number of persons who participated only in ECFE special events or activities, but some district staff questioned whether these counts are as accurate as counts of persons in classes or home visits. CFL staff told us that, in training sessions, they have defined class or home visit participants as those persons who have participated in at least three visits or class sessions during the year. However, this has not been communicated to districts in written instructions for the annual report, so it is unclear whether districts have defined participants in a consistent manner.

ECFE AND SCHOOL READINESS FUNDING ISSUES 47

Table 3.1: ECFE Participation Rates, 1998-99, Selected School Districts

Children in Percentage of Children the District Under Age Five Served in

Under ECFE Classes or District Age Five Home Visits

High Participation Rates:

Fergus Falls 1,048 57.0% Columbia Heights 1,375 52.8 a Pipestone-Jasper 519 50.9 Grand Rapids 1,599 48.8 Warroad 500 48.6 Minnetonka 2,945 47.5 Sibley East 520 45.8 Gibbon-Fairfax-Winthrop 505 43.6 Roseau 561 42.8 Rochester 7,200 39.9

Low Participation Rates:

St. Francis 1,940 8.2 St. Cloud 5,957 9.4 Foley 680 9.4 Cass Lake 505 10.3 Braham-Mora-Ogilvie 1,091 10.4 St. Louis Park 2,270 10.5 Eveleth-Gilbert 598 11.0 Cloquet 913 11.1 Wayzata 3,930 11.6 Howard Lake-Waverly-Winsted 657 12.8

Others:

Minneapolis 26,803 13.2 St. Paul 24,075 29.5

STATE 332,486 23.8% NOTE: This table includes only districts with more than 500 children under age five within the district. a Staff in this district told us that the participation rate reported for 1998-99 appears to be inaccurate. The district’s tally of 1999-2000 participation indicated a rate less than half the 1998-99 rate shown here.

SOURCE: Office of the Legislative Auditor’s analysis, based on data reported by districts to the Department of Children, Families, and Learning.

ECFE participation rates vary considerably among individual districts.

In a 1996 report on the ECFE program, the Department of Children, Families, and Learning recommended expanding program funding so that a larger percentage of the eligible population could be served. It said that “many families do not participate because they are experiencing complex and stressful family situations. Resources for more comprehensive services are of particular importance.” 3 Likewise, a recent report by the Early Care and Education Finance Commission said that “ECFE is unable to reach approximately 95,000 qualifying parents due to insufficient funding.” 4 However,

-Neither the Department of Children, Families, and Learning nor other organizations have systematically collected information that could be used to accurately assess the extent of unmet needs for ECFE services statewide.

Some parents may benefit from ongoing, multi-year participation in ECFE while their children are under age five; others may find that a semester of classes is adequate to meet their needs. Thus, without further evidence of unmet needs, it is not clear that a program whose classes and home visits serve one-fourth of eligible families in a given year (plus many other families through special activities) is underserving Minnesota’s population.

Presently, there is no reliable statewide information on the extent to which families interested in ECFE services have been unable to access them. The department did not collect information on local ECFE waiting lists prior to the 1999-2000 school year. For 1999-2000, the department requested districts to report the “estimated number of families on waiting lists,” but it did not specify a consistent method of counting these families. 5 Consequently, we do not think that the department’s information on local waiting lists is a reliable indicator of unmet needs.

There is no reliable statewide information to indicate whether there are unmet needs for ECFE services.

3 Marsha R. Mueller, Immediate Outcomes of Lower-Income Participants in Minnesota’s Universal Access Early Childhood Family Education (St. Paul: Minnesota Department of Children, Families, and Learning, April 1996), 83.

4 Early Care and Education Finance Commission, The Action Plan for Early Care and Education in Minnesota (Minneapolis, November 2000), 13.

5 The department requested this information in districts’ 1999-2000 annual ECFE reports. However, waiting lists likely fluctuate considerably during the school year, and the department did not specify (1) a date during the reporting period that districts should use when reporting the size of their waiting lists, or (2) a method for ensuring that families on waiting lists are still interested in ECFE services.

6 Participants are asked to report their incomes in specified ranges, so it is not possible to determine the exact number of respondents with incomes above or below the state median household income. The number of participants reporting their household income in the ECFE participant survey represented only 36 percent of the families who participated in ECFE classes or home visits statewide in 1998-99, so it is possible that the survey results do not accurately reflect the actual income distribution among participating families.

On a statewide basis, there is no clear evidence that ECFE is underserving families with below-average incomes. Minnesota’s median household income was $47,240 in 1999, and surveys of 1998-99 ECFE participants showed that 40 percent had incomes under $40,000 and 43 percent had incomes over $50,000. 6 There may be instances where ECFE program fees have discouraged lower-income parents from enrolling, but state law requires school districts to establish “reasonable” sliding fee schedules for these programs and to “waive the fee for a participant unable to pay.” 7 Most of the school district fee policies we reviewed clearly emphasized the option of fee waivers or reductions.

ECFE FUNDING ISSUES Background

Early Childhood Family Education started as a pilot program in 1974 with six districts receiving grants totaling $230,000. In 1975, program size doubled and 12 districts received a total of $500,000 in grants. ECFE continued to grow, both in funding and number of programs until 1981, when appropriations were scaled back for the 1982-83 biennium. In 1983, the Legislature changed the ECFE funding mechanism from allocating funds through competitive grants to allocating funds based on the number of children in each district; grants were also provided to districts to ease the transition to the new funding formula. Fiscal year 1986 was the first year ECFE funds were allocated solely through a statewide funding formula. 8 In 1986, districts received ECFE funding at the rate of $79.25 per child under age five. Funding has since increased to the current per-child rate of $115.96 ($ 113.50 of base funding allowance plus $2.46 of additional state aid), but it has not kept up with inflation. 9 Table 3.2 illustrates the per-child funding rate for ECFE since 1986. 10

In fiscal years 2000 and 2001, each school district received $115.96 for each child under age five residing in the district on October 1 of the previous year. 11 The law sets a minimum level of district ECFE funding equivalent to the amount of base funding that would be allocated if a district had 150 children under age five (in fiscal year 2001, this equals 150 multiplied by $113.50, or $17,025). ECFE revenue is slated to increase to $120 per child for fiscal year 2002, which would increase the minimum funding level for individual districts to $18,000 per year.

ECFE AND SCHOOL READINESS FUNDING ISSUES For 2000 and 2001, ECFE programs received about $116 in state funds per eligible child.

7 Minn. Stat. (2000) §124D. 13, subd. 6.

8 Minnesota Department of Children, Families, and Learning, History of Minnesota Early Childhood Family Education, Program Development (Roseville, MN, 1999) and Lois Engstrom, interview by author, Telephone conversation, Roseville, Minnesota, September 18, 2000.

9 In 1986, the Legislature provided funding to serve 20 percent of the eligible population in some capacity. Program participation levels have increased since 1986, while state funding (in constant dollars) has not. Programs are now serving 24 percent of the eligible population through classes or home visits and an additional 20 percent through special activities. Furthermore, in 1995 most districts shifted community education Teachers Retirement Association (TRA) and FICA expenditures from the general fund to program-specific funds such as ECFE and School Readiness. On average, TRA and FICA expenditures now comprise about 9 percent of these programs’ budgets. See Minn. Laws (1995) ch. 3, art. 4, sec. 15, subd. 1b and Minn. Stat. (2000) §123B. 79, subd. 3.

10 The 1997 Legislature provided a one-time appropriation of $2 million to be used for grants to increase services to families with children under one year of age. Minn. Laws (1997), ch. 162, art. 1, sec. 13 and sec. 18, subd. 7.

11 Minn. Stat. (2000) §124D. 135, subd. 1; Minn. Laws (1999), ch. 205, art. 1, sec. 65; and Minn. Laws (2000), ch. 489, art. 1, sec. 26. Districts receive ECFE funding through a combination of state aid and local levy.

Districts generally receive ECFE aid according to the formula outlined in statute. However, some districts receive less than the statutory minimum funding if (1) the Legislature does not appropriate sufficient funding, resulting in a prorating of each district’s allocation, or (2) they “underlevy” for ECFE. If the Legislature does not appropriate enough revenue, CFL prorates the ECFE aid across all districts – including districts that are supposed to receive the statutory minimum funding level. Due to prorating in fiscal year 1999, those districts that received the minimum level of ECFE funding actually received less than the $17,025 required by statute. 12 Similarly, if a district does not levy the entire amount as required in statute, the ECFE aid must be reduced in proportion. 13 In fiscal year 1999, only four districts did not levy the entire amount required by law and thus received reduced ECFE state aid. 14

Reserve Funds

Table 3.2: ECFE Per-Child Funding Allowance, FY 1986-2001

ECFE Funding Inflation-Adjusted Fiscal Year Per Child ECFE Allowance (2000 $)

1986 $ 79.25 $121.80 1987 79.25 119.47 1988 84.50 122.39 1989 84.50 117.02 1990 85.45 113.04 1991 87.75 110.22 1992 96.50 117.66 1993 101.25 119.82 1994 101.25 116.97 1995 101.25 113.66 1996 101.25 110.69 1997 101.25 107.67 1998 111.25 116.46 1999 113.50 116.90 2000 a 115.96 115.96 2001 a 115.96 N/A a Includes additional $2.46 per eligible child in special state ECFE aid.

SOURCE: Office of the Legislative Auditor’s analysis of Department of Children, Families, and Learning data. Adjusted for inflation using the CPI-W. Bureau of Labor Statistics, Consumer Price Index-Urban Wage Earners and Clerical Workers( 2000); http://stats.bls.gov/top20.html; accessed September 4, 2000.

School districts may use ECFE funds only for ECFE programs.

12 In 1999, 68 districts received less than $17,025 of ECFE funding due solely to prorating.

13 Minn. Stat. (2000) §124D. 135, subds. 3-4.

14 Bird Island, Campbell, Detroit Lakes, and South Koochiching school districts did not fully levy for ECFE in fiscal year 1999.

As outlined in Minnesota statutes, school districts are required to use ECFE funds only for ECFE programming. In addition, all ECFE revenues received by the school district, including state aid, levies, fees, and grants, must be maintained in a reserve account for ECFE. 15 By placing these funds in a reserve account, they are protected from being used for any other school district program. We found that:

-In fiscal year 1999, many school districts had large ECFE reserve fund balances.

In fiscal year 1999, over 20 percent of districts had ECFE reserve fund balances equal to 100 percent or more of their annual state ECFE revenues. Reserve funds from these districts totaled almost $4 million in fiscal year 1999. About 43 percent of districts had fund balances exceeding 50 percent of annual revenues, totaling over $7 million in fiscal year 1999. ECFE program staff at CFL do not systematically monitor districts’ ECFE reserve fund balances. However, in fiscal year 2000, CFL issued guidelines suggesting that districts maintain 8 to 17 percent of annual ECFE revenue in their reserve fund. 16 In 1999, two-thirds of districts had reserve funds larger than 17 percent of annual revenue; reserve funds for these districts totaled over $12 million. Table 3.3 shows total ECFE revenues and reserve fund balances for fiscal year 1999.

To better understand districts with large ECFE fund balances (equal to 100 percent or more of annual revenues), we investigated how these districts differed from those with smaller reserve balances. Table 3.4 shows the average number of ECFE parent-child classes and participants for districts with small eligible populations (150 or fewer ECFE-eligible children). For the most part, the data in Table 3.4 indicate that there are some real differences between small districts with large fund balances and small districts with small fund balances. Specifically, the data in Table 3.4 illustrate that:

Table 3.3: ECFE Revenues and Reserve Balances, FY 1999

Total ECFE Percentage Reserve Total ECFE Districts with of Fund Balances Revenues Reserves Equal to: Number All Districts (in Millions) (in Millions)

Over 100 percent of annual ECFE revenues 78 23% $ 3.9 $ 2.3

Over 50 percent of annual ECFE revenues 146 43 7.1 7.2

Over 17 percent of annual ECFE revenues 231 67 12.1 24.8

ALL DISTRICTS 343 100% $12.2 $37.7 SOURCE: Office of the Legislative Auditor’s analysis of Department of Children, Families, and Learning data.

Almost one-fourth of districts have ECFE fund balances greater than their annual ECFE revenue.

15 Minn. Stat. (2000) §124D. 135, subds. 5 and 7.

16 Betty Cooke, Debbykay Peterson, and Lois Engstrom, Early Childhood Family Initiatives Team, Minnesota Department of Children, Families, and Learning, to Early Childhood Family Education, School Readiness, and Way To Grow Coordinators, Early Childhood Family Education and School Readiness Conference Information, General Information, and Program Updates, March 15, 2000, memorandum.

-In fiscal year 1999, ECFE programs in districts with 150 or fewer ECFE-eligible children and large fund balances served fewer total participants and offered fewer classes than similar-sized ECFE programs with smaller fund balances.

For instance, small districts with large fund balances offered half as many parent-child classes as comparably-sized districts with smaller fund balances. Likewise, small districts with large fund balances served less than half as many children as comparably-sized districts with smaller fund balances. 17

Table 3.5 shows the median percentage of the ECFE-eligible population served by districts of varying sizes and with varying fund balances. The data in Table 3.5 illustrate the importance of district size on a district’s ECFE program. As is clear from the table:

-Regardless of fund balance size, ECFE programs in districts with 150 or fewer eligible children typically served a larger percentage of the eligible population than ECFE programs in districts with over 150 eligible children.

Table 3.5 shows that there is little difference among districts of the same size regarding the median percentage of eligible children served. However, there is a noticeable difference when comparing districts with ECFE-eligible populations of 150 or fewer to districts with larger eligible populations. In fiscal year 1999, the median small district served about 40 percent of the eligible population through classes or home visits and 67 percent in some capacity. At the same time, the median district with more than 150 eligible children served 26 percent of the eligible population through classes or home visits and 45 percent in some capacity. In sum, districts with 150 or fewer eligible children served a significantly larger portion of the eligible population than did larger districts.

Table 3.4: ECFE Parent-Child Classes and Participants in Districts with 150 or Fewer ECFEEligible Children, FY 1999

Large Fund Small Fund Balance Balance N= 29 N= 43

Average number of parent-child classes 15.8 31.9 Average number of participants under age five 70.2 144.2 Average number of parent participants 74.0 86.7

NOTE: “Large fund balances” are defined as reserve funds equal to 100 percent or more of annual ECFE revenues. “Small fund balances” are defined as reserve funds equal to less than 100 percent of annual ECFE revenues.

SOURCE: Office of the Legislative Auditor’s analysis of Department of Children, Families, and Learning data.

Among small districts, those with large ECFE fund balances served fewer families than those with smaller balances.

17 Our analysis indicates that districts with 150 or fewer ECFE-eligible children and small fund balances offered more parent-child classes per capita than did similar-sized districts with larger fund balances. Likewise, these small districts with small fund balances served more parents and children per capita than similar-sized districts with larger fund balances.

In our view, the Legislature should consider possible changes in the ECFE funding formula to address the large reserve fund balances that many districts have. We recommend that the Legislature consider the following policy options:

RECOMMENDATION

The Legislature should consider (1) restricting ECFE funding for districts whose reserve fund balances exceed a certain level; (2) eliminating or reducing the minimum funding level in the ECFE law; or (3) allocating a portion of ECFE funding based on the actual number of persons served, not just the size of the eligible population.

The first option would place a cap on ECFE reserve fund balances and adjust annual allocations to reflect this limit. If limits were placed on ECFE reserve balances, funding that would otherwise go to the capped districts could be redirected to districts with smaller fund balances or higher demand for services. We think it was appropriate for the Department of Children, Families, and Learning to issue guidelines to districts on suitable reserve levels, but incorporating reserve fund policies directly into the ECFE funding formula might encourage greater compliance. This option would require the department to increase its oversight of district fund balances.

ECFE AND SCHOOL READINESS FUNDING ISSUES

Table 3.5: Median Percentage of ECFE-Eligible Population Served, FY 1999

Districts with Districts with 150 or Fewer More Than 150 All ECFE-Eligible Children ECFE-Eligible Children Districts Large Fund Small Fund Large Fund Small Fund

Balance Balance Balance Balance N= 29 N= 43 N= 42 N= 193 N= 307 Median percentage of eligible children served in classes or home visits 39% 42% 25% 26% 28%

Median total percentage of eligible children served in classes, home visits, or special activities. 68 67 39 47 49

NOTE: “Large fund balances” are defined as reserve funds equal to 100 percent or more of annual ECFE revenues. “Small fund balances” are defined as reserve funds equal to less than 100 percent of annual ECFE revenues.

SOURCE: Office of the Legislative Auditor’s analysis of Department of Children, Families, and Learning data.

Small districts served a larger percentage of the ECFE-eligible population than did larger districts.

The Legislature should consider several options for revising the ECFE funding formula.

Limits on districts’ reserve funds may conflict with local school district policies. For example, one district with an ECFE fund balance equal to 42 percent of annual revenues told us that district policy requires a fund balance equal to at least 30 percent of annual expenditures. Another district with a fund balance equal to 26 percent of annual revenues told us that it is building new space for the ECFE program, which will be available in 2003; the district is maintaining a large fund balance in anticipation of the transition to the new facilities.

The second option would eliminate or reduce the minimum ECFE funding level guaranteed in statute. 18 This provision of law was supposed to ensure that the state’s smallest districts could provide at least a basic ECFE program for their district. However, the data indicate that 29 of the 72 programs that receive the minimum ECFE funding level have accumulated at least one year’s worth of funding in reserve while also typically serving a larger portion of their eligible population than most districts in the state. The other 43 districts receiving the state’s minimum funding also generally serve above-average percentages of their residents in ECFE, and only 17 of these districts had fund balances in fiscal year 1999 that were less than 17 percent of their annual revenues. In sum, the state’s smallest districts have typically had high participation rates in their programs, often combined with the accumulation of large fund balances. Staff in one small district with a large ECFE reserve balance told us that the district has used ECFE-dedicated funds for non-ECFE purposes, although we did not independently confirm this. Staff in several other small districts told us they are having difficulty recruiting additional families to participate in ECFE – mainly due to lack of parent interest, not lack of funding. On the other hand, it is possible that reducing the minimum funding level could cause hardship for those small districts that do not have large ECFE reserve fund balances.

Districts that receive the minimum guaranteed level of ECFE funding typically have high participation rates and fund balances larger than recommended by CFL.

18 In fiscal year 1999, districts with less than 150 ECFE-eligible children received a total of $607,000 in additional ECFE funds as a result of the guaranteed minimum funding level.

The third option would allocate a portion of ECFE funding based on the number of children and parents served, rather than allocating all funding based on the number of eligible participants. Funding programs based on population served would likely limit the amount of reserve funds districts could accrue since funding would be linked to services provided. While this would be a change from the current ECFE funding formula, it would more closely parallel Minnesota’s K-12 funding system. 19 In addition, funding ECFE based on population served would provide an incentive for programs to enroll families.

In 1984, the Minnesota Council on Quality Education issued a report on ECFE funding options. The council said that, over the long term, policy makers should consider allocating ECFE funds on the basis of measures other than the number of eligible children in each district. It suggested eventually changing the funding mechanism to a formula based on actual participation. Specifically, the report stated, “In the judgment of the Council, [funding based on eligible population] should be allowed for a five year maximum in the case of a new program. After that, a district’s aid should be linked to actual participation.” 20

Adoption of a participant-based funding formula would require the Legislature or CFL to determine what constitutes a program participant for funding purposes. For example, guidelines would have to be established on how, if at all, to count the participation of a family that attends only one class or special activity. Allocating funding based on population served would also require CFL to more diligently monitor district-reported participation levels. The participation data are not audited for accuracy by CFL, and some districts told us they questioned the accuracy of their reported levels of participation in ECFE special activities. Finally, the Legislature or CFL would have to determine a new per-child funding rate to fairly compensate districts for the ECFE programs provided.

SCHOOL READINESS PROGRAM AVAILABILITY AND PARTICIPATION

The Legislature started the School Readiness program in 1991, and the number of participating school districts has grown considerably since then. In fact,

-All but six of Minnesota’s school districts operated state-funded School Readiness programs in 1999-2000.

ECFE AND SCHOOL READINESS FUNDING ISSUES Funding ECFE based on population served would require increased monitoring of district-reported participation levels.

19 The primary component of Minnesota’s K-12 funding formula, the basic education revenue, equals the per-pupil allowance multiplied by the actual pupil units in each district for the school year. For more information, see Minnesota School Finance, A Guide for Legislators (St. Paul: Minnesota House of Representatives Research Department, December 1998), 20.

20 Council on Quality Education, Study of Alternate Funding Formulas for Early Childhood and Family Education (St. Paul, February 1984), 45.

21 Franconia and Prinsburg are classified as “non-operating” districts, which means they do not operate public schools.

The only districts that did not participate in School Readiness were Farmington, Lyle, Pine Point, New Richland-Hartland-Ellendale-Geneva, Franconia, and Prinsburg. 21 Together, these districts reported 1,795 children under age five within their boundaries in 1998-99. Department of Children, Families, and Learning staff told us that all districts with School Readiness programs have at least one program site within their boundaries, even in cases where the district operates a joint program with another district.

Districts report overall levels of School Readiness participation to the department using the following definition: “A participant is any parent and/or 3½-or 4-year-old child who has had at least one contact with the School Readiness program beyond the mandated basic Early Childhood Screening.” 22 Districts also report information on the number of children who received at least 30 hours of services during the school year. Using these data, we found that:

-Statewide, 32 percent of the state’s children ages three and four participated in School Readiness in 1998-99; 17 percent received at least 30 hours of School Readiness services during the year.

As with ECFE, district-reported School Readiness participation levels varied considerably. Table 3.6 shows large districts (with more than 500 residents under age five) that served high and low percentages of children in School Readiness during 1998-99. 23 We talked with staff in some of these districts and heard various explanations for the participation levels. For instance, staff in the Howard Lake-Waverly-Winsted district said that they counted as participants all families that received a School Readiness newsletter, resulting in the 91 percent participation rate shown in the table. Likewise, one reason for St. Francis’ high rate was that the district’s School Readiness program provided staff to briefly consult with parents while their children underwent mandatory preschool screening— and these contacts were counted as School Readiness participation. In contrast, Wayzata School Readiness staff said that their district had below-average participation in School Readiness because it focused its resources on providing full-year services for a limited number of children. Staff in Osseo said that transportation was a major barrier to family participation in School Readiness, so the district’s low participation rates partly reflect its decision to restrict the number of participants and improve the transportation services provided to them.

Some state officials think that School Readiness services have not been available to enough families statewide. In a 1999 evaluation of the School Readiness program, staff from the Department of Children, Families, and Learning recommended expansion of the program. The report said that:

Children and their families participating in [School Readiness] are typically the ‘working poor’ who do not meet Head Start guidelines, who would not otherwise be involved in a preschool experience, and who may exhibit needs identified during Early Childhood Screening. [School Readiness] is clearly meeting a previously unmet need. Consequently, more of the same is recommended. 24

In 1998-99, about one-half of School Readiness participants received at least 30 hours of service during the year.

22 Department of Children, Families, and Learning, School Readiness 1998-99 Annual Report form, ED-02117-06.

23 We compared the number of all School Readiness participants with the number of three-and four-year-olds in the district. School Readiness is targeted to children older than 3½, but it is possible that some of the participants in 1998-99 were younger than age three.

24 Marsha R. Mueller, Minnesota’s Learning Readiness: 1997-98 Evaluation (Roseville, MN: Minnesota Department of Children, Families, and Learning, January 1999), 16.

Table 3.6: School Readiness Participation Rates, 1998-99, Selected School Districts

Percentage of Children Ages Three and Four Served in Children in 30 or More the District Any School Hours of School Ages Three Readiness Readiness District and Four Activity Activities

High Participation Rates:

Howard Lake-Waverly-Winsted 304 90.8% 32.9% Waseca 371 74.7 19.4 Burnsville 2,201 74.3 43.4 Monticello 565 73.1 15.2 St. Francis 905 72.7 24.3 Watertown-Mayer 282 70.6 6.7 Pine River-Backus 200 66.0 41.0 Warroad 192 65.6 65.6 Grand Rapids 638 65.4 22.6 Stewartville 250 65.2 65.2

Low Participation Rates:

Farmington 701 0.7 0.7 Edina 976 4.0 * Osseo 3,670 5.5 * St. Cloud 2,269 5.8 1.9 Cannon Falls 218 6.4 6.4 Cloquet 399 7.0 7.0 St. Louis Park 908 7.3 6.7 Brainerd 1,107 7.5 7.1 White Bear Lake 1,350 9.0 8.6 Wayzata 1,572 9.5 2.2

Others:

Minneapolis 10,057 15.8 7.9 St. Paul 9,679 31.7 12.6

STATE 137,930 31.6% 16.7% NOTE: This table includes only districts with more than 500 children under age five within the district. The districts shown had the highest and lowest participation rates, based on participation in at least one activity during the year. We excluded three districts that reported overall participation rates exceeding 100 percent of the district’s number of three-and four-year-olds.

*Data for Edina and Osseo are not shown because the number of children the district reported with more than 30 hours of School Readiness services exceeded the total number of children reported participating in any activity.

SOURCE: Office of the Legislative Auditor’s analysis, based on data reported by districts to the Department of Children, Families, and Learning.

However, while we agree that the School Readiness program has helped to fill gaps that would otherwise exist in Minnesota’s system of prekindergarten services,

-The Department of Children, Families, and Learning has not systematically documented the extent to which families interested in School Readiness services have been unable to access them.

The department does not collect information on School Readiness waiting lists or other measures of unmet needs. Data presented earlier indicate that school districts vary in School Readiness participation levels, but the reasons for this variation are not always apparent. In general, it is difficult to quantify the levels of School Readiness services that are being provided and objectively evaluate whether these service levels are sufficient to meet the state’s needs. 25

It is also difficult to determine whether School Readiness services have been sufficiently available to low-income families. State law allows (but does not require) school districts to charge fees for School Readiness, using a sliding scale. The law says that the fees must be set at a level that enables “eligible children of all socioeconomic levels to participate in the program.” 26 The law also requires districts to waive fees for participants unable to pay. It is possible that districts’ fees are prohibitive for some families, but this could not be determined without an in-depth review of the way the fee schedules have actually been applied. In 1998-99, districts reported that 26 percent of parents in School Readiness programs paid the full fees charged by districts, 15 percent paid reduced fees, and 59 percent paid no fees. 27 CFL’s statewide information systems do not contain reliable data on the total fee revenues collected by districts, nor does CFL have information on the rates charged by each district.

In 1998-99, over one-half of families participating in School Readiness paid no fees.

25 State law requires individual School Readiness programs to develop a comprehensive plan “to anticipate and meet the needs of participating families.” There is no comparable statewide plan or needs assessment.

26 Minn. Stat. (2000) §124D. 15, subd. 12.

27 In surveys of School Readiness participants by CFL, 70 percent of respondents during 1998-99 reported household incomes of $50,000 or less, and the state median household income in 1999 was $47,240. However, the number of participants reporting their household income in the voluntary School Readiness participant survey represented only 28 percent of the parents who participated in School Readiness in 1998-99. Thus, it is possible that the survey results do not accurately reflect the actual income distribution among participating families.

We reviewed the 1999-2000 fee schedules of 41 districts. Eleven of these districts charged no fees for their ongoing School Readiness programs. At the other extreme, the Rosemount-Apple Valley-Eagan School District charged $130 per month to families with a child in School Readiness preschool three days per week.

SCHOOL READINESS FUNDING ISSUES

In recent years, many states have implemented (or expanded) publicly-funded prekindergarten programs other than Head Start. A recent Children’s Defense Fund study examined prekindergarten programs that focus primarily on the education of children, and it found that:

-Minnesota (with its School Readiness program) is one of 36 states plus the District of Columbia that have state-funded prekindergarten programs other than Head Start. 28

Three states have “universal” preschool programs; that is, they are not specifically targeted to at-risk children or children from low-income families. Georgia’s program, funded with more than $200 million per year in lottery proceeds, is open to all four-year-olds in the state, and funding has generally been sufficient to accommodate all families wishing to participate. Two other states (New York and Oklahoma) and the District of Columbia also have universal eligibility for their preschool programs, but participation has been limited by funding levels. Minnesota is 1 of 14 states that allow children to enroll in prekindergarten programs (specifically, School Readiness) before they are four years old. 29

In Chapter 2, we noted that Minnesota’s state spending for Head Start was larger than that of most states. In contrast, state spending for Minnesota’s other “kindergarten readiness” program (School Readiness) has not been as generous, compared with similar programs in other states. We found that:

-In 1998-99, Minnesota spent $32 per state resident under age five on the School Readiness program, ranking 24 among the 50 states in spending for public preschool programs other than Head Start.

Heading the list were Georgia ($ 374 per resident under age five), Massachusetts ($ 200), Connecticut ($ 179), Oklahoma ($ 157), Illinois ($ 155), Kentucky ($ 153), and Texas ($ 143). 30 Minnesota’s combined spending for School Readiness and Head Start ($ 90 per state resident under age five) was the fourteenth highest among states.

Minnesota’s funding for School Readiness is modest compared with some states’ funding for similar programs.

28 Karen Schulman, Helen Blank, and Danielle Ewen, Seeds of Success: State Prekindergarten Initiatives, 1998-1999 (Washington, D. C.: Children’s Defense Fund, 1999), 30-31 and 179-209. This includes states that have kindergarten programs for four-year-olds.

29 Ibid., xiii, 27-36, and 59-69.

30 Spending data are from Schulman and others, Seeds of Success, 30-31; July 1999 estimates of state residents under age five are from the U.S. Census Bureau (http://www.census.gov/population/estimates/state/st-99-09.txt; accessed November 1, 2000). The spending data are for child-focused programs only and exclude parent-focused programs such as ECFE. In 1998-99, the ECFE state appropriation per Minnesota resident under age five (not including local levies) was $43.85.

The Minnesota Legislature first appropriated funding for School Readiness (then called Learning Readiness) for the 1991-92 program year, and 365 of the 425 districts (86 percent) participated in School Readiness that year. 31 The Legislature intended the initial $8 million in state funding to be used for start-up costs of the program and did not establish ongoing funding until fiscal year 1994. 32 Funding has increased slightly since 1994, although it has remained relatively flat once adjusted for inflation. Table 3.7 illustrates the funding history for School Readiness through fiscal year 2001.

Table 3.7: School Readiness Funding, FY 1992– 2001

Total School Adjusted for Readiness Funding Inflation (2000 $) Fiscal Year (in Millions) (in Millions)

1992 $ 8.0 $ 9.8 1993 0.0 0.0 1994 9.5 11.0 1995 a 11.0 12.4 1996 9.5 10.4 1997 9.5 10.1 1998 b 10.2 10.7 1999 10.4 10.7 2000 10.4 10.4 2001 10.4 N/A

a In fiscal year 1995, a $1.5 million grant (allocated to all participating districts) was added to the School Readiness grant as part of violence prevention initiatives. This is included in the 1995 funding level reported in this table.

b In fiscal year 1998, $80,000 in one-year grants was given to Greenway and Fergus Falls school districts for special purposes. This is not included in the 1998 funding level reported in this table.

SOURCE: Minnesota Department of Children, Families, and Learning, Minnesota Public Schools Overview of Learning Readiness Program Participation, Funding, and Services January 1992-January 1999( Roseville, MN, 1999). Adjusted for inflation using the CPI-W. Bureau of Labor Statistics, Consumer Price Index-Urban Wage Earners and Clerical Workers( 2000); http://stats.bls.gov/top20.html; accessed September 4, 2000.

School Readiness funding has remained relatively flat, once adjusted for inflation.

31 Minnesota Department of Children, Families, and Learning, “1995 Update: Learning Readiness in Minnesota Public Schools;” http://www.cyfc.umn.edu/children/learnready.html; accessed June 2, 2000.

32 The Legislature allocated $20 million for School Readiness for 1993, but the governor vetoed this appropriation.

Minnesota allocates School Readiness funds to school districts based 50 percent on the number of four-year-old children in the district and 50 percent on the number of pupils in the district eligible for free or reduced-price school lunch. Unlike the ECFE funding formula, there is no minimum funding level for School Readiness programs. As a result, in fiscal year 1999, 134 districts received less than $10,000 in School Readiness funding and 56 districts received less than $5,000 (the lowest funding for a program was $1,200). Because local or statewide changes in the number of four-year olds or the number of children receiving free and reduced-price lunch can change School Readiness aid allocations, state and district staff told us it is difficult to predict the actual funding available for the program. In fact, a firm budget number is generally not available to districts until after the program year has concluded. 33

Minnesota statutes require school districts to use School Readiness funds only for School Readiness programming and transportation. State law also requires districts to keep School Readiness funding in a separate account. 34 However, unlike ECFE, there is no statewide reserve fund for this purpose. As a result, there is no way to verify that districts are maintaining the integrity of School Readiness funds. CFL staff told us that establishing a reserve fund for School Readiness would be useful and would guarantee the preservation of School Readiness funding for the intended programs. Establishing another reserve fund might increase the administrative burden for districts and CFL. However, it would be reasonable to expect CFL to monitor districts’ School Readiness fund balances on an ongoing basis, particularly in light of the large balances some districts have accumulated in their ECFE programs. Establishing a School Readiness reserve fund would provide an additional level of program oversight for School Readiness.

RECOMMENDATION

The Legislature should consider establishing a reserve fund for School Readiness – similar to that established for ECFE.

STATE MONITORING OF DISTRICT FINANCES

Expenditures

School districts use the Uniform Financial Accounting and Reporting Standards (UFARS) computer system to report a variety of financial data, including ECFE and School Readiness program expenditures and account balances. Although questions about the accuracy of UFARS data have been raised in the past, this remains the primary tool that state officials have to monitor school district spending. 35

We reviewed fiscal year 1999 UFARS expenditure data, by district, for School Readiness and ECFE. Several districts reported no expenditures, or very few expenditures, for either School Readiness or ECFE, despite receiving program revenue and providing the services. For example, 15 districts reported no School

ECFE AND SCHOOL READINESS FUNDING ISSUES Several districts reported no expenditures for ECFE or School Readiness although they had received revenue earmarked for the program.

33 Greg Sogaard, Program Finance – Education Finance and Debbykay Peterson, Early Childhood Family Initiatives, Minnesota Department of Children, Families, and Learning, to School Readiness Coordinators and Community Education Directors, School Readiness Aid Entitlement, September 10, 1999, memorandum.

34 Minn. Stat. (2000) §124D. 16, subds. 3-4.

35 For example, see Office of the Legislative Auditor, School District Spending (St. Paul, February 1990).

Readiness expenditures, even though they received School Readiness revenue ranging from $1,800 to $63,700. An additional 33 districts reported spending less than 50 percent of their School Readiness revenue in fiscal year 1999. One district reported only $339 in 1999 School Readiness expenditures, even though the district received over $1 million in School Readiness funding.

CFL staff could not explain the discrepancies between the expenditure and revenue data for ECFE or School Readiness. Staff who administer these programs do not make substantial use of UFARS data to monitor ECFE or School Readiness expenditures. In addition, CFL’s early childhood program staff have not worked with the agency’s finance and management staff to systematically monitor the financial status of districts’ programs. As a result, CFL staff may be missing opportunities to oversee district spending practices and perhaps improve local programs. For example, if CFL finance and management staff regularly provided CFL program staff with information on districts that have spent less than 50 percent of their School Readiness funding in a given year, program staff might help these districts develop strategies to recruit more families or offer more services. In this and other ways, we found that:

-The Department of Children, Families, and Learning provides little oversight of districts’ ECFE and School Readiness expenditures.

Staff who administer the ECFE and School Readiness programs rely largely on personal contacts and districts’ annual reports to obtain information on ECFE program operations. However, districts are not required by law to submit ECFE annual reports (11 of the 346 participating districts did not submit them in 1998-99), and these reports do not include program expenditure information. Districts are required by law to submit biennial School Readiness plan updates, and CFL staff review the content of these documents. These updates include forecasted budget estimates, but they do not provide actual expenditure information.

In our view, CFL program staff should take further advantage of the data already collected by the department through the UFARS computer system. Our comparisons of district expenditure and revenue data raise questions about the consistency and reliability of UFARS expenditure data. However, additional oversight from CFL’s ECFE and School Readiness staff would likely improve the quality and use of the UFARS data.

RECOMMENDATION

The Department of Children, Families, and Learning early childhood program staff should use the data collected through UFARS to monitor districts’ ECFE and School Readiness expenditures and programs.

Reserve Fund Balances

In fiscal year 2000, as noted earlier, CFL staff published guidelines suggesting that districts maintain an ECFE reserve fund balance equal to between 8 and 17 percent of their annual ECFE revenues. Two-thirds of districts had an ECFE fund balance in fiscal year 1999 that exceeded 17 percent of annual ECFE revenues. Large reserve funds have only recently come to the attention of CFL staff, and there is little oversight of these balances. CFL asked districts to report reserve fund balance data in their fiscal year 2000 ECFE annual reports – the first time the department has requested this information in the annual reports. However, CFL already collects this data through UFARS. We spoke with a number of districts regarding their ECFE reserve fund balances over the course of this evaluation. Districts consistently confirmed the accuracy of the reserve fund data contained within the UFARS system. In our view, ECFE program staff could use the existing UFARS data on an ongoing basis to identify districts with large reserve balances and work with them to better use their ECFE funding.

CFL program staff should make further use of the spending data collected through UFARS.

While many districts have large ECFE fund balances, we found that a number of other districts have negative ECFE fund balances. CFL’s Uniform Financial Accounting and Reporting Standards manual states that the ECFE “reserve account is not allowed to go into deficit.” 36 Despite these UFARS guidelines:

-About 25 percent of all districts with an ECFE program had a negative ECFE fund balance during at least one of the three years we examined (fiscal years 1997, 1998, and 1999).

In addition, 24 districts had a negative ECFE fund balance in all three years we examined (1997 through 1999). Among the 346 districts that offered an ECFE program in 1999, 46 had a negative ECFE fund balance. 37 Half of these 46 districts experienced a worsening fund balance between fiscal years 1997 and 1999.

Our discussions with district staff indicated that many districts are unaware that the UFARS guidelines do not allow negative ECFE fund balances. Furthermore, ECFE program staff at CFL were not familiar with the UFARS guidelines that discourage negative ECFE fund balances. CFL program staff indicated that having a negative ECFE balance is an acceptable occurrence and that districts allocate community or general education funding to cover any negative balance. While a negative fund balance is most likely the result of faulty bookkeeping, it could be an indication of a more serious problem. As with the large fund balances, data on negative fund balances are available through the UFARS reporting system and should be used by the ECFE program staff at CFL.

To address problems with both large and negative ECFE fund balances, we recommend the following:

Many districts have negative ECFE fund balances.

36 Minnesota Department of Children, Families, and Learning, UFARS Manual, Balance Sheet Accounts (Roseville, MN, July 2000), 18-19.

37 Districts are treated separately for financial purposes, regardless of whether they shared programs or administrative responsibilities with other districts.

RECOMMENDATION

The Department of Children, Families, and Learning should use the data collected through UFARS to monitor districts’ ECFE fund balances and work with districts to resolve any problems.

Fee Schedules

As discussed earlier, school districts may establish a sliding fee scale for School Readiness programs and are required by law to have a sliding fee scale for ECFE programs. 38 For both of these programs, fees must be waived for any participant unable to pay. In general, we found that:

-There is little oversight of districts’ ECFE and School Readiness fee schedules.

Currently, CFL program staff collect data on forecasted School Readiness fee revenues through districts’ biennial plan updates; they do not have reliable information on actual fee revenues collected. CFL also does not collect any information on the levels of ECFE program fees paid by participants. Without additional data on fee revenues, CFL cannot get an accurate picture of how much total revenue districts have available for ECFE or School Readiness programs. In addition, without information on variation in districts’ fee revenues, it is difficult to assess whether districts’ fees pay for portions of total program costs that are too large or too small. We surveyed 41 districts across the state to learn about their ECFE and School Readiness fee schedules and found a large range in fees charged. However, 6 of the 41 districts reported that they charge no fees for ECFE, despite the requirement in law that every district establish an ECFE sliding fee scale. CFL program staff should work with such districts to create an appropriate fee scale.

RECOMMENDATION

The Department of Children, Families, and Learning should monitor districts’ ECFE and School Readiness fee schedules to ensure compliance with state law. The department should also collect accurate information on the total fee revenues actually collected for these programs.

38 Minn. Stat. (2000) §124D. 15, subd. 12 and Minn. Stat. (2000) §124D. 13, subd. 6.

4 Program Outcomes and Oversight

SUMMARY

Existing research indicates that well-implemented early childhood programs can provide educational benefits to children— although the evidence is mixed regarding whether these benefits persist over time. Claims of widespread program successes should be viewed skeptically because early childhood programs differ considerably from site to site, and studies have not conclusively identified which program components produce good outcomes. Presently, Minnesota’s statewide education information systems cannot readily track the progress of children from Head Start, Early Childhood Family Education (ECFE), or School Readiness programs who have entered the K-12 education system. Also, federal and state officials regularly conduct on-site reviews of individual Head Start programs, but external reviews of local ECFE and School Readiness programs are not required and have been done on a very limited basis.

Previous chapters of this report described the participants in Minnesota’s early childhood programs, the services provided, and the way these programs are funded. Ultimately, however, it is important to consider the impacts these programs have on their participants. This chapter addresses the following questions:

-What has research nationally and in Minnesota shown about the outcomes of early childhood education programs? Does research provide clear guidance about when interventions should start, how long they should last, or the best types of services?

-How, if at all, has Minnesota measured the progress of children who have participated in Head Start, Early Childhood Family Education (ECFE), and School Readiness?

-To what extent do state, federal, and other agencies monitor local early childhood education programs on an ongoing basis?

To answer these questions, we examined the findings of existing research on early childhood programs in Minnesota and elsewhere. To examine practices for overseeing early childhood programs, we interviewed state, federal, and other officials, reviewed information reported to state officials by service providers, and obtained external monitoring reports.

GENERAL RESEARCH FINDINGS

Most research on early childhood education programs in the U.S. has been conducted during the past 40 years. This section briefly summarizes what research has indicated about program outcomes and the timing of early childhood interventions.

Program Effects

Early childhood education programs take a variety of forms but, in general, they are intended to enhance children’s cognitive, social, emotional, and physical development. Usually these programs try, at a minimum, to lay a foundation for subsequent success in school. Many program advocates have suggested that early interventions might also help disadvantaged children lead more productive lives— through higher earnings, less criminal behavior, and improved health. We found that:

-Researchers generally agree that well-implemented, high-quality preschool programs can help children, at least in the short term.

Some of the strongest evidence has come from “model” early education programs implemented during the past 40 years. Model programs have usually been small, generously funded, and carefully implemented, so they are not necessarily typical of large-scale programs operating elsewhere. Nevertheless, it is noteworthy that a summary of research from model programs concluded that:

[Model programs] demonstrate unequivocably that quality preschool programs can provide an immediate boost to children’s intellectual performance and reduce their rate of placement in special education classes. The studies also provide moderate evidence that quality preschool programs decrease grade retention and increase the likelihood of high school graduation. 1

Likewise, RAND researchers reviewed model programs and reported that “in some situations, carefully targeted early childhood interventions can yield measurable benefits.” 2 RAND concluded that long-term cost savings to government outweighed program costs in the two programs targeting at-risk children for which the researchers could reliably make such comparisons. 3

“Model” early education programs have demonstrated some strong, positive effects.

1 Ron Haskins, “Beyond Metaphor: The Efficacy of Early Childhood Education,” American Psychologist 44, n. 2 (February 1989), 276.

2 Lynn A. Karoly, Peter W. Greenwood, Susan S. Everingham, Jill Hoube, M. Rebecca Kilburn, C. Peter Rydell, Matthew Sanders, and James Chiesa, Investing in Our Children: What We Know and Don’t Know About the Costs and Benefits of Early Childhood Interventions (Santa Monica, CA: RAND, 1998), xiii. One of the nine programs reviewed by RAND was a large-scale program, not a model program.

3 Ibid, 73-103. One of the studies involved a preschool for children ages three and four; the other involved home visits to families with children age two and under. The savings included increased tax revenues and reduced expenditures for welfare, criminal justice, education, health, and other services.

Table 4.1 presents excerpts from several sources that have summarized research literature on early childhood programs’ impacts on children. The summaries by Haskins and Barnett note that studies of ongoing, large-scale early education programs, such as Head Start, have usually found smaller impacts than have studies of model programs. For instance, the widely-heralded outcomes of the Perry Preschool Program in Michigan occurred in a model program that looked quite different from traditional Head Start or other preschool programs. 4 Compared with most other preschool programs, the Perry program had more favorable staff-to-child ratios, better educated teachers, more frequent home visits, and better technical assistance for staff.

In addition, -Research has reported mixed findings regarding the extent to which the benefits of preschool programs have been sustained in the years following program completion.

Many studies of model and large-scale programs have reported that early education interventions have positive initial benefits for children, including increased IQ and improved school achievement. Often, however, the IQ gains of participants erode within a few years— contrary to some of the claims made by early advocates of these programs. Studies of early childhood programs have reported somewhat more promising program results using measures other than IQ. For example, some studies have linked program participation with long-term school success, using measures such as achievement tests, grade retention rates, and special education placement rates. Even with these measures, however, the research findings are mixed. 5 In cases where long-term effects have faded over time, some researchers have partly attributed this to the low quality of grade schools attended by the participants. 6

In addition, -Studies have not provided very conclusive evidence about the characteristics of effective early childhood programs.

RAND’s recent summary of research said: “The two most important things we don’t know about early childhood intervention programs is why the successful programs work— and why those not shown to be successful don’t.” 7 It noted, for instance, that research has not yet revealed clear answers about whether programs should focus services on the child, the parent, or both. 8 The summary also noted that there is limited evidence on how to identify children who would benefit most from interventions. Some researchers have concluded that more intensive early childhood services are likely to produce better results, but the research has not provided clear answers about the preferred amount of service hours per week or the frequency of home visits. 9

PROGRAM OUTCOMES AND OVERSIGHT Initial gains by children sometimes fade within a few years.

4 In the Perry study, 58 children participated in a model preschool program for one or two years, and 65 were in a control group. Based on more than two decades of tracking, researchers found that the program reduced participants’ grade retention, welfare usage, and crime and increased their school completion rates and employment levels. However, as one of the founders of Head Start (Edward Zigler) has observed, the Perry program had “few resemblances to Head Start”— see Zigler and Sally J. Styfco, “Head Start: Criticisms in a Constructive Context,” American Psychologist 49, n. 2 (February 1994), 128.

5 In addition to mixed overall findings in the reports, the findings for particular subgroups of participants have also varied. For instance, four experimental studies of model programs found larger effects on achievement test scores for girls than boys— see W. Steven Barnett, “Long-Term Effects of Early Childhood Programs on Cognitive and School Outcomes,” The Future of Children

5, n. 3 (Winter 1995), 41.

6 Valerie E. Lee and Susanna Loeb, “Where Do Head Start Attendees End Up? One Reason Why Preschool Effects Fade Out,” Educational Evaluation and Policy Analysis 17, n. 1 (Spring 1995), 62-82.

Table 4.1: Impacts of Preschool Programs-- A Sampling of Research Summaries

W. Steven Barnett, Rutgers University: “The weight of the evidence establishes that [early childhood care and education] can produce large effects on IQ during the early childhood years and sizable persistent effects on achievement, grade retention, special education, high school graduation, and socialization. In particular, the evidence for effects on grade retention and special education is overwhelming. Evidence is weaker for persistent achievement effects…. Comparison of estimated long-term effects between model programs and large-scale programs indicates that the latter tend to have smaller effects.” 1

Ron Haskins, staff for U. S House of Representatives Committee on Ways and Means: “Research has shown that both model programs and Head Start have immediate positive impacts on tests of intellectual performance and social competence but that this impact declines over the first few years of public schooling. The evidence of improvement on long-term measures of school performance such as special education placement is substantial for model programs but thin and inconsistent for Head Start. There is limited but provocative evidence that model programs may have positive effects on life success measures such as teen pregnancy, delinquency, welfare use, and employment, but there is virtually no evidence linking Head Start attendance with any of these variables.” 2

Edward Zigler and Sally J. Styfco, Yale University: “Studies of Head Start and other early intervention programs reached the same conclusion: Preschool graduates generally show immediate gains in intelligence and achievement test scores, but these benefits do not appear to be permanent.… The evidence— disappointing to some— is that early intervention cannot guarantee success for life…. Evaluators have focused almost exclusively on the preschool education component and its effects on intelligence and achievement. Possible effects to physical health, nutritional status, social behavior, parents’ child-rearing abilities, family functioning, parental empowerment, and community development have been under-evaluated and undervalued.” 3

U.S. General Accounting Office: “Although an extensive body of literature exists on Head Start, only a small part of this literature is program impact research. This body of research is inadequate for use in drawing conclusions about the impact of the national program in any area in which Head Start provides services such as school readiness or health-related services. Not only is the total number of studies small, but most of the studies focus on cognitive outcomes, leaving such areas as nutrition and health-related outcomes almost completely unevaluated.” 4

Craig T. Ramey and Sharon Landesman Ramey, University of Alabama-Birmingham: “The evidence accumulated over the past 25 years indicates that early intervention programs can produce modest to large effects… on children’s cognitive and social development.... Over time, the initial positive effects of early interventions will diminish to the extent that there are not adequate environmental supports to maintain children’s positive attitudes and behavior and to encourage continued learning related to school.” 5

U.S. Administration on Children, Youth, and Families: “Exposure to Head Start does not usually eliminate developmental disparities between groups of children that originate in the diverse cultural backgrounds of families, though it may reduce them. At the end of the Head Start year, children in the best programs are at national norms for early literacy and math skills, but children in many programs are not.” 6

1 W. Steven Barnett, “Long-Term Effects of Early Childhood Programs on Cognitive and School Outcomes,” The Future of Children5, n. 3 (Winter 1995), 43, 44. 2 Ron Haskins, “Beyond Metaphor: The Efficacy of Early Childhood Education,” American Psychologist44, n. 2 (February 1989), 274.

3 Edward Zigler and Sally J. Styfco, “Head Start: Criticisms in a Constructive Context,” American Psychologist49, n. 2 (February 1994), 128, 129. 4 U.S. General Accounting Office, Head Start: Research Provides Little Information on Impact of Current Program( Washington, D. C., April 1997), 2. 5 Craig T. Ramey and Sharon Landesman Ramey, “Early Intervention and Early Experience,” American Psychologist53, n. 2 (February 1998), 115, 117. The authors define “modest to large” effects as ranging from “0.2 to over 1 standard deviation.” 6 U.S. Department of Health and Human Services, Administration on Children, Youth, and Families, Head Start Program Performance Measures: Second Progress Report (Washington, D. C., June 1998), 47.

In the case of Head Start, a forthcoming national study may shed some light on its effectiveness. In 1998, Congress mandated completion of a national analysis of the impact of Head Start programs by September 2003. 10 The study will examine factors that may affect program impact, such as program quality, a child’s length of time in the program, the child’s age when entering the program, and the intensity of services. Such a study may be useful, but there will always be questions about how to apply the findings from large-scale research to programs that were not direct subjects of study. 11 Early childhood programs often differ considerably from site to site— for instance, in their daily curricula, quality of staff, scope of services, and the skill with which they have been implemented. For this reason,

-The findings of individual early childhood research studies should be applied with considerable caution to programs elsewhere.

PROGRAM OUTCOMES AND OVERSIGHT Early childhood programs may differ considerably from site to site.

7 Karoly and others, Investing in Our Children, 108-109.

8 A National Academy of Sciences report recently said that “programs that offer both a parent and a child component appear to be the most successful in promoting long-term developmental gains for children from low-income families”— see From Neurons to Neighborhoods: The Science of Early Childhood Development, ed. Jack P. Shonkoff and Deborah A. Phillips (Washington, D. C.: National Academy Press, 2000), 345. In contrast, another extensive review of the literature found that “there is no evidence that the type of parent involvement used in past research studies has led to greater benefits for children, more cost-efficient programs, better maintenance of effects, or benefits for other family members”— see Karl R. White, Matthew J. Taylor, and Vanessa D. Moss, “Does Research Support Claims About the Benefits of Involving Parents in Early Intervention Programs?,”

Review of Educational Research 62, n. 1 (Spring 1992), 120.

9 Karoly and others, Investing in Our Children, 109-111; Ramey and Ramey, “Early Intervention and Early Experience,” 115-116; Barbara A. Wasik and Nancy L. Karweit, “Off To a Good Start: Effects of Birth to Three Interventions on Early School Success,” in Preventing Early School Failure: Research, Policy, and Practice, ed. Robert E. Slavin, Karweit, and Wasik (Boston: Allyn and Bacon, 1994), 13-57.

10 Head Start Reauthorization Act, 42 U.S. Code 9844 (g) (1998).

11 Recently, a study of 1970-71 Head Start participants from selected Florida and Colorado sites reported evidence of some long-term program impact, based on a 17-year follow-up— see Sherri Oden, Lawrence J. Schweinhart, and David P. Weikart, Into Adulthood: A Study of the Effects of Head Start (Ypsilanti, MI: High/Scope Educational Research Foundation, 2000). Nearly one-third of Minnesota’s Head Start grantees use in whole or part the type of curriculum used in this study. On the other hand, there have been many changes in Head Start standards and practices since the program examined in this study was in place.

Most of the discussion in this section has focused on preschool-type programs, which often provide direct services to children for many hours each week. One of Minnesota’s main early childhood programs— Early Childhood Family Education (ECFE)— is quite different from this model. ECFE is sometimes referred to as a “two-generation” program because it aims to inform and educate parents— through classes, discussions, or structured parent-child activities— in addition to offering developmentally appropriate activities for children. ECFE classes typically meet for just two hours per week. Two-generation programs can take many forms, but:

-Studies of two-generation programs have generally found small or no effects on child development, although many have reported some positive impact on parenting skills.

Researchers at Abt Associates reviewed six “premier” two-generation programs and reported few measurable impacts on children. 12 Several of the programs had positive effects on parenting— for example, resulting in increased time spent with children, improved teaching skills, and improved parent-child interactions. The Abt researchers said that the intensity of these programs was an important factor in their impact, with “low intensity” programs (such as weekly or biweekly home visits, combined with occasional parent meetings) providing limited opportunity to have an impact. Based on the Abt review and others, a recent summary of research concluded that: “Fairly strong evidence suggests that the best way to promote child development is to work directly with children and not to assume that changes in parents will lead to changes in children.” 13 Likewise, a summary of research by Craig and Sharon Ramey concluded that early childhood programs emphasizing direct educational experiences for children have shown larger and more enduring benefits than programs that rely on indirect ways to improve child competencies (such as parent training). 14

Even if the outcomes of two-generation programs have not been dramatic, it is possible that parent-focused services might be one important element of broad, multi-faceted strategies to support families with young children. Still, the relatively weak child impacts found in studies of parent-focused programs may indicate that programs such as ECFE should not, by themselves, be viewed as a substitute for more intensive, child-focused services— particularly in the case of children considered to be at risk for later school problems.

Parent education may be one component of an effective service strategy.

But parent focused services alone have shown limited impacts on children.

12 Robert G. St. Pierre, Jean I. Layzer, and Helen V. Barnes, “Two-Generation Programs: Design, Cost, and Short-Term Effectiveness,” The Future of Children 5, n. 3 (Winter 1995), 76-93. The studies also indicated that two-generation programs increased the rate of participation by children and their parents in relevant social and educational services, and they found that programs had large variations in outcomes among individual sites.

13 Deanna S. Gomby, Mary B. Lerner, Carol S. Stevenson, Eugene M. Lewit, and Richard E. Behrman, “Long-Term Outcomes of Early Childhood Programs: Analysis and Recommendations,”

The Future of Children 5, n. 3 (Winter 1995), 16.

14 Ramey and Ramey, “Early Intervention and Early Experience,” 116. Based on such findings, the authors said, “The practical question for the field of early intervention, however, is whether parent education and general family support programs can be justified if they do not produce child benefits.”

Timing of Program Interventions

Policy makers are interested not only in whether early interventions are effective in general, but also in which specific strategies work. They would like to know when to intervene with young children, whether to have special follow-up services in grade school for preschool graduates, and how long early interventions should last. In general,

-Research has provided only limited guidance about the “right” age to intervene with at-risk children and how long to intervene, although there is general agreement that it would be unrealistic to expect brief interventions to have lifelong impacts.

Service providers have tried many program options to prevent children’s failure in school— including various interventions in preschool, kindergarten, and early elementary grades. Few studies have compared these options, but one found that short-term preschool programs were not enough, by themselves, to prevent early school failure. It reported that the program with the strongest results through third grade offered a variety of interventions from prekindergarten through elementary years— including high quality preschool, full-day kindergarten, enhanced curriculum and instruction, tutoring, parent involvement, family support programs, and small classes. 15

Another study that addressed the issue of program duration was the Abecedarian Project, a model program involving 111 at-risk children in North Carolina. This study looked at the impact of a long-term intervention that, for some study participants, started shortly after birth. It found that eight-year-olds who received continuous intervention for their first eight years of life academically outperformed children who only participated in preschool and those who only received three years of assistance in elementary grades. 16 Other analysts observed, however, that the study was not able to disentangle whether the program’s positive outcomes were attributable to its duration, its intervention at infancy, or some other aspect of the way the program was delivered. 17 Regarding the grade school portion of this program, some researchers have questioned the effectiveness of school-age interventions that are intended to help prolong the positive effects of preschool programs. A forthcoming, federally-sponsored study found that programs specifically designed to help Head Start children make the transition into regular schools had little measurable impact on children’s academic and social development. 18

In recent years, there has been considerable discussion about the need for early childhood programs for children under age three. Researchers continue to debate what the evidence shows about intervention with infants. Researchers who worked on the Abecedarian Project noted that “five major studies demonstrating some of the largest effects of early intervention on children’s early cognitive and social development… all enrolled children during infancy.” 19 However, one recent literature review said that “evidence about when programs should begin and how long they should last is mixed,” 20 and another described the issue of when to intervene as an “unanswered” issue. 21 More generally, the National Academy of Sciences recently said that “the research literature on service intensity, duration, and age of initiation is perhaps the most complex and inconclusive aspect of the early childhood intervention knowledge base.” 22 It noted that evidence has shown earlier interventions to be more justifiable for some specific conditions— such as hearing loss, vision impairment, and major delays in motor skill development— than for others. 23

Researchers continue to debate the merits of services for children under age three.

15 Robert E. Slavin, “Preventing Early School Failure: Implications for Policy and Practice,” in Preventing Early School Failure: Research, Policy, and Practice, 206-229. This study found that “the most effective [strategies] by far for preventing early reading failure are approaches incorporating one-to-one tutoring of at-risk first-graders” (p. 212).

16 Frances A. Campbell and Craig T. Ramey, “Effects of Early Intervention on Intellectual and Academic Achievement: A Follow-Up Study of Children From Low-Income Families,” Child Development 65, 684-698.

17 For example, see John T. Bruer, The Myth of the First Three Years: A New Understanding of Early Brain Development and Lifelong Learning (New York: The Free Press, 1999), 169-170.

18 Mr. Tom Schultz, U.S. Head Start Bureau, interview by author, Telephone conversation, St. Paul, September 27, 2000. In addition, Barnett, “Long-Term Effects,” reported that, “The only direct support for the need for prolonged, schoolage intervention comes from a single quasiexperimental study,” but he said that improvements in poor-quality schools might independently improve child development (p. 44).

19 Ramey and Ramey, “Early Intervention and Early Experience,” 115.

Some advocates for expanded birth-to-three services have cited recent research on brain development— suggesting, for example, that early stimulation causes brain synapses to form, or that missed opportunities to help children’s brains develop before age three will adversely affect them for the rest of their lives. We found that:

-For the most part, brain research does not offer clear evidence about the right time to begin programmatic interventions in young children’s lives or the types of care and instruction that should be provided.

Most brain researchers agree that early experiences are important and help to build the foundation for subsequent brain development. On the other hand, they generally do not believe that brain research provides much insight into how to raise children or accelerate brain development. As one recent review of the research concluded:

The new scientific research doesn’t say that parents should provide special “enriching” experiences to [babies] over and above what they experience in everyday life. It does suggest, though, that a radically deprived environment could cause damage. 24

Some people regard the first three years of life as the critical period for brain development— and, thus, the time when interventions should occur in order to maximize their impact. As noted above, this appears to be true for some children— such as those with serious disabilities or those subject to neglect or abuse. However, two recent research summaries suggest that the birth-to-three years are not the only time for effective interventions with most children:

First, …most learning is not subject to critical-period constraints, not confined to windows of opportunity that slam shut. Second, neuroscientists do not think that the quantity of experience or stimulation during the critical period is the key variable in brain development. Third, critical periods are complex. There are distinct phases within critical periods; there are distinct critical periods for specific functions within a system like vision or language; and the periods and phases within them extend over considerable periods of time— years— and well into children’s second decade of life. Finally, critical periods do not all fit neatly into the first three years of life. 25

Research has not clearly indicated how to encourage brain development in very young children.

20 Barnett, “Long-Term Effects,” 44.

21 Karoly and others, Investing in Our Children, 109.

22 From Neurons to Neighborhoods, 362.

23 Ibid., 216, 364.

24 Alison Gopnik, Andrew N. Meltzoff, and Patricia K. Kuhl, The Scientist in the Crib: Minds, Brains, and How Children Learn (New York: William Morrow and Company, 1999), 182.

Available evidence indicates that such critical periods are more exceptional than typical in human development…. Assertions that the die has been cast by the time the child enters school are not supported by neuroscience evidence and can create unwarranted pessimism about the potential efficacy of interventions that are initiated after the preschool years. 26

Overall, early childhood education research provides limited guidance regarding the design of effective early childhood education programs. The research results are complex and sometimes contradictory. In general, we think that policy makers should be cautious and even skeptical about claims of widespread, long-term impacts demonstrated in previous research. It is hard to predict how long it will take for the research evidence to provide clearer insights, but for now policy makers will have to continue making important decisions about program funding and design based on limited knowledge of program results.

MINNESOTA STUDIES OF PROGRAM OUTCOMES

ECFE Evaluations

Over the past 25 years, there have been many studies of Minnesota’s ECFE program by state agencies and private consultants. Before the Legislature authorized statewide implementation of the program in 1984, reports generally praised the quality of the programs that had been implemented. 27 For example, a consultant offered the following comments on 13 pilot projects:

On the whole, these are outstanding demonstration programs— carefully conceived, diligently implemented and enthusiastically received by program participants. As full-time evaluators, we see a large variety of programs…. It is unusual to find programs where implementation so closely follows intent, and where program implementation has been so thorough, so intense, and so successful. 28

In surveys initiated by state agencies during the program’s pilot phase (before 1984), parents generally expressed satisfaction with ECFE. More recently, state evaluators have used surveys and interviews mainly to help assess the impact of the ECFE program on parents. For example, based on standardized interviews of 183 parents at the beginning and end of their ECFE programs, a 1992 report concluded that “the majority of coded parent responses reflected parent change in feelings, knowledge, expectations, and behavior” after participating in ECFE. 29 Examples of these changes included increased feelings of support from others and increased confidence and self-esteem as parents. In a 1996 report, a large majority of 239 low-income parents from 14 school districts reported that ECFE participation improved their understanding of child development and their relationships with their children. 30

There is limited evidence about the components of effective early education programs.

25 Bruer, The Myth of the First Three Years, 103.

26 From Neurons to Neighborhoods, 216.

27 For a summary of pilot program evaluations, see Minnesota Department of Education,

Evaluation Study of Early Childhood Family Education: Report to the Legislature (St. Paul, March 1, 1986).

28 Michael Q. Patton, An External Review of Early Childhood and Family Education Pilot Programs (Minneapolis: Minnesota Center for Social Research, March 1977), 2, 4.

There have been several efforts to evaluate ECFE programs using measures other than parent perceptions. Most recently, a 1996 study of ECFE programs in 14 school districts found that ECFE staff gave fewer participating parents low ratings on various parenting measures at the end of the school year than at the beginning— based on analyses of parent interviews. 31 For instance, staff rated 39 percent of participating parents as having low knowledge and awareness of their children at the beginning of the year, compared with 17 percent at the end of the year. In this same study, independent raters who viewed videotapes of parent-child interactions gave 8 percent of the parents higher ratings at the end of the year than at the beginning. 32

Overall, we conclude that: -Department of Children, Families, and Learning staff have demonstrated ongoing interest in evaluating the ECFE program. However, these evaluations have not provided definitive evidence of ECFE’s impact on parents or children.

The studies of ECFE have been ambitious and time-consuming, and they have involved close working relationships between state and local ECFE staff. Department staff indicated that the ECFE evaluations have helped state and local staff identify program improvements and plan staff training activities. Staff from the department raised money from private sources to pay for the most recent study, which received a national award as an exemplary evaluation. 33 Nevertheless, the limitations of the various ECFE studies are notable.

ECFE evaluations have helped state staff identify program improvements and plan training activities.

29 Betty Cooke, Changing Times, Changing Families: Minnesota Early Childhood Family Education Parent Outcome Interview Study (St. Paul: Minnesota Department of Education, March 1992), ix.

30 Marsha R. Mueller, Immediate Outcomes of Lower-Income Participants in Minnesota’s Universal Access Early Childhood Family Education (St. Paul: Minnesota Department of Children, Families, and Learning, April 1996), 60.

31 Mueller, Immediate Outcomes, 42, 64. For example, staff rated parent behavior by asking parents how they maintained self-control when they were frustrated with their children, how they guided their children’s behaviors, and how they responded to an angry child.

32 Ibid., 75. In 1979, a survey of about 50 kindergarten teachers indicated that most perceived that children from ECFE programs had better attitudes toward school, better school-readiness, and better relationships with their parents than other kindergarten children— see Michael Q. Patton, The Council on Quality Education Administered Early Childhood Family Education Program: A Perspective on Impact (Minneapolis: Minnesota Center for Social Research, 1979). In 1980, teachers and independent observers in two school districts rated the characteristics of ECFE participants and non-participants when they were in kindergarten; they did not find measurable differences— see Anderson and Berdie Associates, Inc., Early Childhood and Family Education: A Program Outcome Assessment (St. Paul: Minnesota Council on Quality Education, December 1980).

33 The evaluation was funded with a $150,000 McKnight grant and a $20,000 legislative appropriation. It received an exemplary evaluation award from the American Evaluation Association.

First, the only sure way to attribute parent changes to the ECFE program would be through the use of a “control group” study that carefully compares groups of randomly-selected ECFE participants and non-participants. The two most recent ECFE studies (1992, 1996) reported that parents experienced changes in their feelings, knowledge, expectations, and behavior during the year. But such changes might be typical of people gaining experience as parents, talking with other parents, and reading child-rearing books. The changes could also have been caused by participation in other early childhood programs, such as Head Start or private preschools. Our recommendations on program monitoring (later in this chapter) do not include implementation of control group studies, but we think that future evaluations by the department should acknowledge the possible impact of factors other than the ECFE program on child development.

In addition, the ECFE studies have not tested whether the improvements they reported in parent ratings were large enough to be statistically significant. 34 Furthermore, none of the ECFE parent surveys or interviews conducted over the past 25 years have been based on samples that could be presumed to be representative of participants statewide.

It is difficult to determine whether parenting improvements reported in previous studies were caused by ECFE or other factors.

34 The editor of an evaluation journal expressed concern about this absence of standards for judging the size of the changes reported in the 1996 study— see Jody L. Fitzpatrick, “Dialogue With Marsha Mueller,” American Journal of Evaluation 19 (1), 1998, 97-98. The Immediate Outcomes report said that “designing the evaluation to test statistical significance would have required a research design that would not have been as useful to program staff and their questions even though it would have been useful to research and policy communities” (p. 92).

Finally, the studies have provided little evidence of effects on parenting or school readiness subsequent to the end of the ECFE program year. A 1986 report said that the most effective way to measure ECFE outcomes would be to track differences in participants and non-participants over a 10-to 20-year period, and it outlined a series of tasks that would be needed to monitor program quality and costs. 35 Likewise, the 1996 ECFE evaluation said: “Behavior change is often a long-term proposition requiring support, practice, and reflection.” 36 A full evaluation of such changes would require a longer follow-up than the ECFE studies have provided.

School Readiness Evaluations

The Minnesota Department of Children, Families, and Learning has issued several evaluation reports on the School Readiness program since 1995. 37 In a 1999 study of 177 children in School Readiness programs, a majority of the children were rated as “proficient” at the end of the school year on various measures of child development. 38 The study also found that most of the children rated as not proficient at the beginning of the year had improved by the end of the year. In addition, kindergarten teachers said that 66 percent of the former School Readiness participants were “doing well” in kindergarten, 27 percent were “making adequate adjustment” to kindergarten, and 7 percent were experiencing a “difficult” adjustment. The report also said that 99 percent of interviewed parents described changes their children experienced during the school year “as a result of their [School Readiness] experience.” 39

Again, as with ECFE, we think the department has demonstrated ongoing interest in evaluating the School Readiness program. However, we think it is also important to convey to policy makers that:

-Evaluations of the School Readiness program have not provided definitive evidence of its effects.

The department has sometimes declared that School Readiness has a positive impact without acknowledging other possible explanations for the results of its evaluations. According to the department, “Completed studies show young children benefit from participation in the School Readiness program.” 40 However, the progress shown by children during the school year on developmental checklists might be explained by normal maturation or factors other than the School Readiness program. Without a study that compares a control group of School Readiness participants with a group of non-participants, it would be difficult to conclude whether the program has had an impact. The department’s evaluations do not discuss the possibility that factors other than the School Readiness program may have contributed to the children’s development or to parents’ perceptions of improvement.

35 Minnesota Department of Education, Evaluation Study of Early Childhood Family Education, 2, 27-29.

36 Mueller, Immediate Outcomes, 76.

37 Early reports included Marsha R. Mueller, Learning Readiness: Parent and Teacher Perspectives About Minnesota’s Learning Readiness Programs— A Report of Findings From the 1994 Pilot Evaluation (St. Paul: Minnesota Department of Education, January 1995) and Marsha R. Mueller, Learning Readiness: 1996-1997 Evaluation Summary (Roseville, MN: Minnesota Department of Children, Families, and Learning, February 1998).

38 Marsha R. Mueller, Minnesota’s Learning Readiness: 1997-98 Evaluation (Roseville, MN: Minnesota Department of Children, Families, and Learning, January 1999), 4. On 16 measures of personal and social development, the percentage of children rated as proficient ranged from 55 (“ uses words to resolve conflicts”) to 90 (“ uses classroom materials purposefully and respectfully”). On 11 measures of literacy and language, the percentage of children rated as proficient ranged from 65 (“ recognizes the association between spoken and written words”) to 89 (“ listens with interest to stories read aloud”). The sample consisted of children from eight school districts.

39 Ibid., 10. Seventy-four parents were interviewed.

40 Minnesota Department of Children, Families, and Learning, Evaluation of Minnesota’s School Readiness Program (Formerly Learning Readiness), 1995-2001 (Roseville, MN, undated), 1.

Head Start Evaluations

Head Start grantees regularly report information to federal and state agencies on the characteristics of children they serve and the services they provide. This information was the basis for much of our discussion of the Head Start program in Chapter 1. But, in contrast to ECFE and School Readiness,

-No large-scale studies have examined the effects of Minnesota’s Head Start program on its participants.

This may reflect the limited goal of Minnesota’s Head Start subsidy program: to expand the availability of the federal Head Start program to additional Minnesota families. Also, some program staff told us that Head Start’s effectiveness was sufficiently established in national studies, thus minimizing the need for state-specific studies. As discussed earlier in this chapter, however, national research on Head Start has shown mixed results, and findings from research elsewhere may not necessarily apply to Minnesota programs due to site-to-site program differences.

Presently, individual Head Start grantees are considering ways to measure children’s progress while in the program. A 1998 federal law required each federal Head Start grantee serving children ages three to five to create a system to track child and program outcomes. 41 Grantees must fully implement this tracking system no later than the 2001-02 program year, and they will be expected to include child outcome information in annual self-assessment reports. Grantees have been instructed to collect “some data” about each of eight areas of child learning and development, but the federal government has not prescribed which measures grantees should use or set standards for acceptable levels of performance. 42 Grantees are only required to collect information about children’s progress during the time they are attending Head Start, so these new outcome tracking systems will not necessarily yield insights into the success of children’s transition into the regular school system.

In addition to these efforts, the Minnesota Department of Children, Families, and Learning is working with consultants to identify outcome measures for state-funded Head Start services for children under age three.

A new federal law will require Head Start agencies to track children’s progress while they are in the program.

41 Head Start Reauthorization Act, 42 U.S. Code 9836a (b)( 2)( B) (1998).

42 U.S. Department of Health and Human Services, Administration on Children, Youth, and Families, to Head Start Grantees and Delegate Agencies, Using Child Outcomes in Program Self-Assessment, August 10, 2000, memorandum.

ONGOING STATE OVERSIGHT

One way to improve program quality and accountability is to periodically conduct evaluations of program results, such as those discussed in the previous sections of this chapter. Another approach is through ongoing oversight of the programs offered by individual service providers. In the following sections, we discuss the extent to which the Department of Children, Families, and Learning (CFL) and others have monitored local service delivery practices.

In general, the department oversees the ECFE and School Readiness programs somewhat differently than it oversees Head Start. Department staff told us that they prefer to let local school districts have primary responsibility for overseeing and supervising district programs, with CFL providing policy leadership and technical assistance to improve program quality. In contrast, department staff told us that they have traditionally played a more active oversight role with grantees other than school districts— such as agencies that receive state and federal Head Start grants.

Monitoring Local ECFE and School Readiness Programs

-The Department of Children, Families, and Learning provides ongoing training and technical assistance to local ECFE and School Readiness programs, but it has not conducted regular, on-site reviews of these programs.

State staff help local school districts implement ECFE and School Readiness programs. For instance, when the Legislature authorized statewide development of ECFE programs in the early 1980s, state officials prepared a “best practices” guidebook to help school districts. 43 Recently, CFL staff prepared a resource guide with benchmarks that local program staff can use to evaluate the social, intellectual, physical, and emotional development of four-year-olds. 44

In addition, state law requires districts to bienially submit summaries of their School Readiness programs to CFL, including program descriptions and lists of cooperative arrangements with other service providers. 45 Districts are not eligible for state aid until this “plan update” has been approved by the CFL commissioner. School districts are not required to submit such reports for their ECFE programs.

CFL asks each school district to annually submit data on the number of early childhood program participants and the types of services provided. 46 For ECFE, districts report information on the number of participants in (1) classes and home visits, and (2) special activities. 47 For School Readiness, districts report the total number of participants during the year, as well as the number who participated for at least 30 hours during the year. Districts keep attendance records but are not required to report to CFL how many times during the year each participant attended an ongoing activity, such as a parenting class or a daily preschool.

CFL monitors individual Head Start programs more actively than it does ECFE or School Readiness programs.

43 Minnesota Council on Quality Education, A Guide for Developing Early Childhood Family Education Programs (St. Paul, January 1984).

44 Department of Children, Families, and Learning, Minnesota Early Childhood Indicators of Progress: A Resource Guide (Roseville, MN, 2000).

45 Minn. Stat. (2000) §124D. 16, subd. 1.

46 For the 1998-99 program year, CFL received reports from 97 percent of ECFE programs and 96 percent of School Readiness programs.

CFL staff also request districts to collect demographic information on all ECFE and School Readiness participants, using a participant questionnaire. Statewide, however, a majority of participants did not complete questionnaires during the 1998-99 program year, so it is unclear whether the data collected by districts on ECFE and School Readiness participants are representative of all participants in these programs. 48

Recently a report by the Children’s Defense Fund identified Minnesota as 1 of 21 states that has “insufficient monitoring” of local prekindergarten programs. 49 It cited Minnesota’s lack of regular, on-site reviews as the basis for this judgment, and it said that:

No matter how comprehensive a prekindergarten initiative’s standards are, states cannot ensure that programs achieve positive results without regularly scheduled monitoring visits. Monitoring enables states to identify programs that are not maintaining an adequate level of quality. 50

The Children’s Defense Fund report noted that 15 states monitor their public prekindergarten programs (other than Head Start) on site on a regular basis. Some of these states specify how often programs must be reviewed (for example, six states require annual visits); other states schedule on-site reviews on a less regular schedule.

Minnesota law does not require CFL to conduct on-site reviews of ECFE and School Readiness programs, and staffing constraints presently limit the agency’s ability to do more on-site monitoring. CFL has less than 2.0 full-time-equivalent professional staff assigned to ECFE and School Readiness. In our view, state officials should monitor local ECFE and School Readiness programs more directly than they do now, but it would be unrealistic for CFL to do this without additional staff.

Although CFL does not conduct regular on-site program reviews, a small number of districts have arranged for peer reviews of their programs. The Minnesota Community Education Association has a long-standing process by which districts may request peer reviews of their community education programs. The requesting district can determine the scope of the review, but nearly all such reviews examine ECFE programs and many also examine School Readiness programs. Typically, staff from other school districts spend two days reviewing documents and conducting interviews. The studies generally look at organizational structure, administrative procedures, strategic planning, community involvement practices, program services, collaborative efforts, and facility use. Statewide, however, an average of only five such reviews were conducted annually during the past decade. Twenty districts have undergone two or three peer reviews since 1979, while most districts were never reviewed during this period. 51

CFL should do more on-site monitoring of ECFE and School Readiness, but it has less than 2.0 FTE staff assigned to these programs.

47 Districts are supposed to report unduplicated counts of participants in these two categories, but some district staff told us that they have not ensured that double-counting between these categories does not occur.

48 In the 1998-99 program year, districts received completed parent questionnaires from 29 percent of School Readiness participants and 36 percent of ECFE participants in classes or home visits.

49 Karen Schulman, Helen Blank, and Danielle Ewen, Seeds of Success: State Prekindergarten Initiatives, 1998-1999 (Washington, D. C.: Children’s Defense Fund, 1999), 121-123. For Minnesota, the report focused on the School Readiness program.

50 Ibid., 121.

In 1998, CFL staff began developing a “program enhancement process” for ECFE, with the intent of increasing the number of on-site program reviews beyond those conducted by the Minnesota Community Education Association. CFL officials hope that this process will enable each district to have a peer review every four years. CFL will present details of this process to local ECFE program coordinators statewide in Fall 2001.

A final mechanism for ongoing program oversight is local advisory councils. State law requires school boards to establish advisory councils to help oversee ECFE and School Readiness programs. A majority of ECFE council members must be participating parents, and they are supposed to help the school board in “developing, planning, and monitoring” the ECFE program. 52 The law specifies that School Readiness advisory councils contain a mixture of service providers and parents, and the council must “monitor the progress of the program” and advise the school board on administrative matters. 53 School Readiness advisory councils also must review and approve local School Readiness plans and plan updates. We did not examine the activities of local advisory councils in our study.

Overall, we think that some additional state oversight of ECFE and School Readiness would be helpful— to more effectively monitor the finances of local programs (discussed in Chapter 3) and to monitor compliance with state program standards or “best practices.” 54 Such monitoring might improve quality in locally-administered programs and provide greater accountability for state expenditures.

On average, only five school districts per year have had peer reviews of their early education programs.

51 For instance, Minnesota Community Education Association records dating to 1979 indicate that there have been no reviews of the programs in the state’s largest school district (Minneapolis). The most recent reviews for some other large districts include St. Paul (1980), Duluth (1985), Rochester (1983), and Anoka (1979).

52 Minn. Stat. (2000) §124D. 13, subd. 9.

53 Minn. Stat. (2000) §124D. 15, subd. 7.

54 CFL has issued guidance to districts that describes ECFE program expectations in areas such as parent education, administration, budgeting, staffing, and linkage with other programs.

RECOMMENDATION

The Department of Children, Families, and Learning should consider the need for additional staff to oversee the ECFE and School Readiness programs— through internal reallocation or through a legislative request.

Monitoring Local Head Start Programs

The federal government has adopted performance standards for Head Start services, and agencies receiving federal funds are expected to comply with these standards. To ensure compliance,

-Minnesota Head Start grantees receive annual on-site monitoring reviews. Over a three-year period, each grantee is reviewed once jointly by federal and state staff and twice by state staff only.

Federal reviewers observe Head Start services and facilities on site, and they interview staff responsible for administrative, fiscal, health, nutrition, and educational services. Minnesota is 1 of only 12 states in which state-level Head Start staff participate in the federal reviews or conduct their own reviews. 55 When federal reviewers find instances of non-compliance, they require the grantee to submit a letter outlining how the problem will be corrected. If reviewers find a more serious problem— referred to as a “deficiency”— they require grantees to submit a quality improvement plan within 30 days, and subsequent reviews are more stringent. Unresolved deficiencies can result in termination or sanction of a grantee; a 1999 study by the U.S. Department of Health and Human Services reported that over 100 grantees nationwide have lost their funding since 1993. 56 No Head Start grantees in Minnesota have been terminated.

We examined 33 reports on individual Minnesota Head Start grantees that federal officials issued between 1996 and 2000, based on triennial monitoring visits. Nearly every triennial review noted some sort of compliance issue— such as improper placement of a fire alarm, lack of individualization in the curriculum, missing records, or inadequate financial monitoring systems. Five of the federal reviews found deficiencies, and the grantees were asked to develop quality improvement plans.

We also reviewed 31 reports of Minnesota grantees that resulted from monitoring visits by state officials only. These reviews were less comprehensive and formal than the joint federal-state reviews. For instance, state staff examined a limited number of compliance issues in these reviews, and they used the reviews partly to explore program concerns with grantees (such as difficulties finding dentists to provide services to children in Head Start). State officials told us that these reviews are an important, ongoing means of identifying program improvements.

Minnesota is 1 of 12 states in which state staff help review Head Start grantees.

55 Schulman and others, Seeds of Success, 123.

56 U.S. Department of Health and Human Services, Advisory Committee on Head Start Research and Evaluation, Evaluating Head Start: A Recommended Framework for Studying the Impact of the Head Start Program (Washington, D. C., 1999), 7. Many other grantees received intensive technical assistance to help them return to compliance.

We found that federal and state regulators collect a wealth of information about Head Start participants and services. Federal Head Start performance standards require grantees to annually submit a comprehensive Program Information Report (PIR). 57 The PIR provides detailed data about the characteristics of families served, program operations, enrollment, staffing, and community outreach. In 1999, grantees provided answers to more than 260 questions on the PIR questionnaire, including information on all individual participants. In general, we think that:

-There is more complete, detailed information about services and participants in Head Start than there is for ECFE or School Readiness.

Finally, it is worth noting that all Head Start grantees are subject to oversight by members of their local communities. Federal standards require each Head Start grantee to have a policy council of parents and community members. The policy council conducts an annual self-assessment of the grantee, and it has authority to make various decisions regarding the grantee’s staffing, services, and enrollment priorities.

Monitoring Long-Term Educational Outcomes

One of the underlying goals of Head Start, ECFE, and School Readiness is to lay a foundation for children’s school success in kindergarten and subsequent grades. We found that:

-The Minnesota Department of Children, Families, and Learning has not systematically tracked the progress of early childhood program graduates who have entered the K-12 school system, although some individual school districts have done so.

In recent years, some Minnesota districts have used the nationally-recognized “Work Sampling System of Child Assessment” to measure child progress. This assessment can be used from preschool through Grade 5 to monitor child development and identify ways to enhance instruction. Presently, 111 districts are using the Work Sampling System to assess children in their School Readiness programs, although state officials have very limited information about specific ways in which the assessments have been implemented and what they have shown. 58

One Minnesota district (Minneapolis) formally assesses all children during the first weeks of kindergarten, and it has specifically examined the performance of children who have participated in various early education programs. Two years ago, the results of these assessments led Minneapolis district officials to urge publicly-funded preschool programs to implement curricula with stronger cognitive elements.

Some school districts regularly assess the progress of children in early childhood programs.

57 U.S. Department of Health and Human Services, Head Start Bureau, Head Start Program Performance Standards and Other Regulations (Washington, D. C., 1996), 186 (45 CFR 1304.51( h)( 2)).

58 For the most part, CFL officials do not know which grade levels or curriculum areas are a part of individual districts’ tracking systems. The assessment can track seven curriculum areas (personal and social development, language and literacy, mathematical thinking, scientific thinking, social studies, the arts, and physical development), but districts using the assessment do not necessarily track progress in all of these areas.

 

In our view, the state should develop the capacity to directly measure the performance of former early childhood program participants in the K-12 school system. It is probably impractical for the state or individual districts to initiate “experimental” studies that would compare the progress of randomly-selected program participants with non-participants. However, it would be useful for state officials to examine trends in grade retention rates, special education placement rates, and standardized test scores among children previously served by Minnesota’s early education programs. Such tracking would not provide definitive evidence of program results, but it might help state and local officials detect emerging issues. 59 Early childhood programs can vary considerably from site to site, so it makes sense for state officials to build the capacity to directly measure child outcomes, rather than presuming that outcomes in Minnesota programs will mirror those found in other studies.

Officials with the Minnesota Department of Children, Families, and Learning told us that they thought it would be valuable to track the performance of Head Start, ECFE, and School Readiness participants once they enter the K-12 school system. This could be done with a retrospective approach— that is, by looking at the early childhood and K-12 records of a group of school-age children who were previously in early childhood programs. Alternatively, it could be done prospectively— by assigning identification numbers to children as they enroll in early childhood programs, which could subsequently be linked with CFL’s K-12 records. CFL officials said that either strategy is technically feasible, but both would require additional resources. In general, we think that the prospective approach has the advantage of allowing CFL to track child progress on an ongoing basis, rather than relying on a one-time study.

RECOMMENDATION

The Legislature should ask the Department of Children, Families, and Learning to determine the steps and resources that would be required to track the elementary school progress of children who participated in Head Start, School Readiness, or ECFE.

The department may be able to provide legislators with information during the 2001 session regarding the merits of child tracking options. If not, the Legislature should require the department to prepare a report on this topic before the next session.

Minnesota should track the K-12 performance of former participants in early childhood programs.

59 For instance, low or declining third grade test scores among School Readiness graduates might indicate a need to explore service improvements for at-risk children— such as more intensive or higher quality preschool services, or better ways of helping children sustain their preschool progress in elementary school. Also, it might be useful to look for differing trends among subgroups of participants.

-The Legislature should consider amending Minn. Stat. §119A. 52 to authorize agencies other than those that received Head Start funding in 1989 to be eligible for innovative grants (p. 35).

-The Legislature should consider amending Minn. Stat. §119A. 52 to authorize the Department of Children, Families, and Learning to determine a grantee’s number of state-funded slots based on its proposed services and the cost per child of those services, rather than based on the federally-set rates per child (p. 37).

-The Legislature should consider increasing the percentage of state Head Start funds allocated on the basis of unmet needs (p. 38).

-The Legislature should consider: (1) allocating state Head Start funding to be used for all eligible children under age six, allowing grantees to determine how to allocate their funding among children of various ages, and (2) broadening the income eligibility guidelines for state-funded Head Start slots (p. 42).

-The Legislature should consider (1) restricting ECFE funding for districts whose reserve fund balances exceed a certain level; (2) eliminating or reducing the minimum funding level in the ECFE law; or (3) allocating a portion of ECFE funding based on the actual number of persons served, not just the size of the eligible population (p. 53).

-The Legislature should consider establishing a reserve fund for School Readiness – similar to that established for ECFE (p. 61).

-The Department of Children, Families, and Learning early childhood program staff should use the data collected through UFARS to monitor districts’ ECFE and School Readiness expenditures and programs (p. 62).

-The Department of Children, Families, and Learning should use the data collected through UFARS to monitor districts’ ECFE fund balances and work with districts to resolve any problems (p. 64).

-The Department of Children, Families, and Learning should monitor districts’ ECFE and School Readiness fee schedules to ensure compliance with state law. The department should also collect accurate information on the total fee revenues actually collected for these programs (p. 64).

Summary of Recommendations

-The Department of Children, Families, and Learning should consider the need for additional staff to oversee the ECFE and School Readiness programs – through internal reallocation or through a legislative request (p. 81).

-The Legislature should ask the Department of Children, Families, and Learning to determine the steps and resources that would be required to track the elementary school progress of children who participated in Head Start, School Readiness, or ECFE (p. 83).

Further Reading

Bruer, John T. The Myth of the First Three Years: A New Understanding of Early Brain Development and Lifelong Learning (New York: The Free Press, 1999).

Early Care and Education Finance Commission. The Action Plan for Early Care and Education in Minnesota: Final Report (Minneapolis, November 2000).

From Neurons to Neighborhoods: The Science of Early Childhood Development,

ed. Jack P. Shonkoff and Deborah A. Phillips (Washington, D. C.: National Academy Press, 2000).

Gopnik, Alison, Andrew N. Meltzoff, and Patricia K. Kuhl. The Scientist in the Crib: Minds, Brains, and How Children Learn (New York: William Morrow and Company, 1999).

Minnesota Department of Children, Families, and Learning. Head Start in Minnesota (Roseville, MN, 2000).

Mueller, Marsha R. Immediate Outcomes of Lower-Income Participants in Minnesota’s Universal Access Early Childhood Family Education (St. Paul: Minnesota Department of Children, Families, and Learning, 1996).

Minnesota’s Learning Readiness: 1997-98 Evaluation (Roseville, MN: Minnesota Department of Children, Families, and Learning, 1999).

Novak, Kathy. Review of State Funding for Head Start (St. Paul: Minnesota House of Representatives Research Department, December 1999).

Oden, Sherri, Lawrence J. Schweinhart, and David P. Weikart. Into Adulthood: A Study of the Effects of Head Start (Ypsilanti, MI: High/Scope Educational Research Foundation, 2000).

Punelli, Danyell. Early Childhood Development and Care Programs (St. Paul: Minnesota House of Representatives Research Department, June 2000).

Schulman, Karen, Helen Blank, and Danielle Ewen. Seeds of Success: State Prekindergarten Initiatives, 1998-1999 (Washington, D. C.: Children’s Defense Fund, 1999).

The Future of Children 5, n. 3, (Winter 1995). (This entire issue is devoted to long-term outcomes of early childhood programs.)

U.S. Department of Health and Human Services, Head Start Bureau. Head Start Program Performance Measures: Second Progress Report (Washington, D. C., 1998).

U.S. General Accounting Office. Head Start: Research Provides Little Information on Impact of Current Program (Washington, D. C., April 1997).

 

January 2, 2001 James Nobles Legislative Auditor Office of the Legislative Auditor Centennial Building 658 Cedar Street St. Paul, Minnesota 55155

Dear Mr. Nobles: The Department of Children, Families & Learning welcomes the opportunity to respond to your December 2000 report entitled Early Childhood Education Programs. We found the report to be an informative and ambitious look at the complexity of Minnesota’s early childhood education programs. We appreciate the intensive effort your staff has put forth to provide an informative perspective on these programs, specifically Head Start, Early Childhood Family Education (ECFE) and School Readiness. We believe the report generally presents a fair picture of the current delivery of services to Minnesota’s young children and their families by the aforementioned programs. The report provides valuable information to guide our work as we look forward to implementing some of the suggested improvements.

We especially appreciate the recommendation regarding tracking the elementary school progress of children who participate in early childhood education programs. The development of the capacity to directly measure the performance of former early childhood program participants in the K-12 school system would be particularly helpful in furthering the school readiness of our youngest citizens. National research has demonstrated that participation in comprehensive early childhood education programs (the Abecedarian program, for example) can lead to immediate gains in cognitive test scores, better kindergarten achievement, lower rates of grade retention and special education placement, and higher rates of high school graduation. Studies have shown that particularly for poor children, high quality early childhood education can have positive effects on scholastic achievement that lasts through high school. While such gains are certainly beneficial to those individual children and should be pursued for this reason, the advantages benefit the larger society as well. It is much more cost effective to provide early childhood education than to remedy matters later. We would assert that the body of research is more convincing than the report purports.

1500 HIGHWAY 36 WEST T: (651)582-8200 ROSEVILLE, MN 55113-4266 TTY: (651)582-8201

http://cfl.state.mn.us

We maintain that the report does not sufficiently recognize the difference in administrative frameworks between Head Start, a grant program, and School Readiness and ECFE, both schoolbased programs. We would further attest that this difference influences levels of oversight and data collection. Both state agency and federal staff directly supervise Head Start grantees, and families who participate in Head Start must meet income eligibility requirements. Consequently, we have considerably more information about participating Head Start families. School Readiness and ECFE are school-based early childhood programs and, as such, are supervised and administered by local school districts. These programs are more autonomous in their program design and are subject to the financial management, auditing and reporting procedures required of public schools.

The report uses averages and percentages as a basis for comparison among programs. While useful and easy to understand, this methodology sometimes falls short of presenting the information necessary to complete the picture. For example, School Readiness and ECFE offer a broad continuum of services to an entire population rather than a prescribed set of services to a targeted population. In this case, the use of averages tends to mask some important details such as the more intensive, more costly services to children and families with the greatest needs. To illustrate, families that participate in family literacy programming through School Readiness and ECFE may spend 20-25 hours or more per week and receive services in addition to literacy programming depending upon identified needs. Strategies such as this contrast greatly in intensity and cost with the special one-time events at the opposite end of the continuum. Both need to be acknowledged, however, to provide a balanced overall picture of these programs.

Similarly, using percentages to compare low-income participation sometimes serves to limit the scope of the issue. For example, as indicated in Table 1.6, a total of 43% of Head Start children are from families with incomes under $10,000, as compared to only 6% of ECFE and 8% of School Readiness children. However, when comparing numbers, this is a total of 5,732 Head Start children compared to 4,710 ECFE children and 3,986 School Readiness children. (The Department acknowledges that these numbers may be duplicative because some children are appropriately involved in more than one program.)

In closing, I want to thank your staff for their professional commitment to this study and for their willingness to go the extra mile to learn about our programs. I know my staff has greatly appreciated their willingness to work cooperatively in the development of this report. Thank you again for this opportunity to respond to the report.

Sincerely,

/s/ Christine Jax, Ph. D.

Christine Jax, Ph. D. Commissioner Department of Children, Families & Learning

Home | Financial Audit Division | Program Evaluation Division