U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Office of Human Development Services Administration for Children, Youth and Families on DOC* HE 23.1002 D33/2/V.1 hds human development services - Final Report of National Day Care Home Study Volume I are amny in the United States: findings National Day Care Home Study UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAi BOOKSTACKS U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Office of Human Development Services Administration for Children, Youth and Families Day Care Division DHHS Publication No. (OHDS) 80-30282 Bd September 1981 Prepared for the Day Care Division • Administration for Children, Youth and Families • Office of Human Development Services • U.S. Department of Health and Human Services Final Report of the National Day Care Home Study Contract No. 105-77-1051 Volume I Family Day Care in the United States: Summary of Findings Principal Author Steven Fosburg Project Director Abt Associates Cambridge, Massachusetts With Contributions By: Judith D. Singer Barbara Dillon Goodson Donna Warner Nancy Irwin Lorelei R. Brush Janet Grasso Editor: Nancy Irwin National Day Care Home Study Digitized by the Internet Archive in 2013 http://archive:org/details/familydaycareinu01fosb_0 L 'OC. VOLUMES IN THE FINAL REPORT SERIES OF THE NATIONAL DAY CARE HOME STUDY • Reports available from the Administration for Children, Youth and Families or from ERIC Document Reproduction Service, P.O. 190, Arlington, Virginia 22210. • Executive Summary (Administration for Children, Youth and Families)— Synopsis of the find- ings from all study components including data on family day care providers, the children in their care, and the children's parents. Presents information on the nature of day care in each of the study settings and presents both cost and program data on family day care systems. • Volume 1, Summary of Findings (Abt Associates Inc.) — Details the issues outlined in the Ex- ecutive Summary. • Volume II, The Research Report (Abt Associates Inc.) — Focuses on the caregiver and the children in her care and presents extensive descriptive and statistical analyses of the interview and observation data collected. It includes profiles of both the caregiver and the children in care, discusses the stability of the day care arrangements, the group composition of the family day care homes, and the costs of providing care. Concludes with a comparative analysis of the observed behaviors of caregivers and the children in their care. • Volume III, Observation Component (SRI International) — Presents the findings from the observations conducted in day care homes in the three study sites (Los Angeles, Philadelphia, and San Antonio) and detailed descriptions of the methodologies used. • Volume IV, Parent Component (Center for Systems and Program Development) — Presents the information provided by the parents of the children in the family day care homes; describes these parents, their needs and preferences for care, and their satisfaction with family day care; and focuses on child day care costs. • Volume V, Family Day Care Systems Report (Abt Associates Inc.) — Presents an extensive descriptive and statistical analysis of the day care institutions that administer family day care systems. These systems are one of the principal mechanisms for providing subsidized day care in a family day care setting, and the cost analyses in this volume are the first attempt to estimate the cost of providing such care. • Volume VI, The Site Case Study Report (Abt Associates Inc.) — Describes the status of family day care in each of the study sites based on interviews with knowledgeable respondents ranging from state licensing staff to day care advocates. This volume is intended to describe the context in which the study was conducted and thereby to provide the reader a fuller understanding of the study findings. • Volume VII, The Field Operations Report (Abt Associates Inc.) — Describes the steps used to implement the study in three study sites. • Family Day Care Manuals. (Abt Associates Inc.) These booklets, designed for day care systems and family day care providers, describe operations, processes and costs of running a family day care home. This Summary of Findings is abstracted from the NDCHS Final Report Series prepared by Abt Associates of Cambridge, Massachusetts, SRI International of Menlo Park, California, and Center for Systems and Program Develop- ment of Washington, D.C. in TABLE OF CONTENTS Page Foreword ix Chapter 1 : INTRODUCTION 1 Chapter 2: DESIGN AND METHODOLOGY 11 Chapter 3 : STATE REGULATION OF FAMILY DAY CARE 25 Chapter 4: CHARACTERISTICS OF FAMILY DAY CARE PROVIDERS 37 Chapter 5: DAY CARE CHILDREN: GROUP COMPOSITION .... 51 Chapter 6: PARENTS OF THE CHILDREN IN CARE 65 Chapter 7: OBSERVATIONS IN FAMILY DAY CARE HOMES ... 73 Chapter 8: COSTS OF FAMILY DAY CARE 95 Chapter 9: FAMILY DAY CARE SYSTEMS 107 Chapter 10: CONCLUSIONS AND RECOMMENDATIONS 123 LIST OF TABLES Page Chapter 1 Table 1: Table 2: Chapter 2 Table 3: Table 4: Table 5: Table 6: Table 7: Chapter 3 "Table 8: Number of Children Aged 0-14 in Care 10 Hours a Week or More, 1975 3 Estimates and Projections of Number (In Thousands) of Children Under Six Years with Mothers in the Labor Force, 1965-1990 4 National Day Care Home Study Design: Ethnicity and Regulatory Status 12 Distribution of Core Caregiver Sample 15 Planned Distribution of Caregiver Interview Sample for the Main Study 16 Distribution of Observation Sample (Main Study) 17 Distribution of Parent Sample (Main Study) 18 Proportion of States with Standards for Certain Aspects of Family Day Care 26 v Chapter 4 Table 9: Table 10: Table 11 Table 12: Table 13: Chapter 5 Table 14: Table 15: Table 16: Table 17: Table 18: Table 19: Table 20: Chapter 7 Table 21: Table 22: Table 23: Chapter 8 Table 24: Table 25: Table 26: Table 27: Table 28: Table 29: Table 30: Table 31 Page Median Caregiver Age 39 Percent of Homes with Any Related Child 41 Percentage of Married Caregivers 12 Median Caregiver Income 43 Median Years of Education 44 Mean Number of Nonresident Children Per Home by Caregiver, Ethnicity and Regulatory Status 53 Number of Nonresident Children Per Home by Site, Caregiver, Ethnicity and Regulatory Status 55 Mean Number of Resident Children Under Seven Per Home 56 Mean Number of Nonresident Related Children Per Home 58 Percentage of Nonresident Related Children 59 Age Distribution of Children in Care by Regulatory Status 60 Distribution of Age Group Configurations 62 Distribution of the Caregiver's Time 77 Distribution of the Caregiver's Time by Regulatory Status 8 1 Caregiver Behaviors Associated with Training 91 Mean Hourly Fee Per Child 96 Mean Weekly Fee Per Child 96 Mean Hourly Fee Per Child by Whether or Not a Relative is in Care 97 Mean Net Hourly Wage 98 Mean Weekly Revenue from Fees by Type of Care 100 Mean Weekly Expenditures per Child for Food 101 Average Weekly Net Income per Home from Caregiving . . 101 Mean Weekly Fee Per Paying Child Under Current and Alternative Minimum Wage Requirements for Providers 104 Chapter 9 Table 32: Profile of 22 Family Day Care Systems 109 VI LIST OF FIGURES Chapter 1 Figure 1 : Chapter 4 Figure 2: Figure 3: Chapter 5 Figure 4: Figure 5: Chapter 7 Figure 6: Chapter 8 Figure 7: Chapter 9 Figure 8: Figure 9: Figure 10: Page Family Day Care 10 Hours a Week or More 7 Distribution of Caregiver Age 39 Distribution of Years of Education 45 Distribution of Number of Children per Home 54 Distribution of Resident Children Under 14 57 Distribution of Caregiver's Time 79 Distribution of Caregiver Wage Rates 99 Functional Breakdown of Hourly Resource Costs .... 117 Core Versus Supplemental Resource Costs 118 Average Functional Costs: Administration, Direct Care and Supplemental Services 119 vu FOREWORD Every day, in every community throughout America, millions of children leave their homes and enter the world of family day care. Who are these children and who are the women who provide their care? What does a family day care home look like? What kinds of care do children receive in family day care homes? What role does family day care play in the community? How does it fit into the overall day care picture? These questions and others provided the impetus for the National Day Care Home Study. They reflect the growing recognition of family day care as an im- portant factor in the lives of American children. At the same time, such ques- tions show a need for the most fundamental, most basic level of knowledge, reflecting the fact that little has been known about family day care or how it functions, even though it is the most widely used form of day care outside the child's own home, serving children from infancy through the elementary school years. Historically, family day care has played an important part in the friendship and kinship networks which form the structure of American society. Informal child care by friends and relatives is an American tradition, as old and as fun- damentally a part of the community as the family itself. More recently, new forms of family day care have appeared as federal and state governments have become involved in day care as a service to working parents. The importance of family day care for families, for communities and for government programs and policies will continue to grow as increasing numbers of women with young children enter the labor force through the end of this century. The Administration for Children Youth and Families operates under a broad mission of advocacy for children and their families. The Day Care Divi- sion is concerned in particular with finding ways to support parents, day care providers, program administrators, state agencies and others in their child care endeavors. An important aspect of this advocacy role is the sponsorship of research to provide valid and reliable information from which to derive a sound framework for program planning and policy decisions. The National Day Care Home Study was designed to establish such a foundation for family day care. Initiated in 1976, the National Day Care Home Study was a four-year study of urban family day care in three major American Cities — Los Angeles, Philadelphia and San Antonio. The principal goal of this research was to understand the many dimensions of family day care as it operates in different ix geographic, cultural and structural settings for children of different ages. It is the first large-scale study of family day care in this country and the first to study all of the major participants — children and their parents, caregivers, program administrators, agency officials and advocates in the day care com- munity. It is thus the first study to approach family day care from an ecological point of view as a comprehensive and complex phenomenon inter- woven throughout the fabric of American life. We appreciate the hard work of the many people who together have made this study a success. Dr. Preston Bruce, Jr., ACYF Day Care Division Director at the time the study was conducted, provided administrative direction to the project, facilitating the many difficult and complex tasks faced in the research. We very much appreciate the hard work and wisdom of the three contractors who jointly conducted the study: Abt Associates Inc., SRI International and the Center for Systems and Program Development. Our consultant panel was also extremely helpful and generous with their expertise. Finally, we are deeply grateful to the study participants— the parents, caregivers, children and the many other individuals who gave so freely of their knowledge, their wisdom and their time. Without them, the study could not have been conducted. The Administration for Children, Youth and Families is proud to present these findings from the National Day Care Home Study. We are confident that the increased understanding of family day care afforded by this research will be translated into programs and policies to enhance the quality of family day care and provide needed assistance to parents and providers, states and local communities, as we strive together for the benefit of American children. Patricia Divine-Hawkins Project Director, NDCHS Day Care Division Administration for Children, Youth & Families August, 1981 Washington, D.C. Chapter 1 INTRODUCTION The National Day Care Home Study was a four-year study of urban family day care sponsored by the Day Care Division of The Administration for Children, Youth and Families in the Office of Human Development Services, Department of Health and Human Services. Family day care is nonresidential child care provided in a private home other than the child's own. Encompassing a myriad of unique arrangements between families and their day care providers, family day care constitutes the largest, most complex system of child care in the United States. It is distinguished from care in a day care center, nursery school or other group facility as well as from substitute care provided in the child's own home. Family day care is composed of three major categories delineated by regulatory and administrative structure. The largest of these categories con- sists of unregulated providers who operate informally and independently of any regulatory system. A second category consists of regulated (licensed or registered) caregivers who meet state and/or federal standards but, except for this link with the broader day care community, operate independently. The third, and smallest, group of providers consists of regulated homes which are operated as part of day care systems or networks under the administrative auspices of a sponsoring agency. "" Nearly half of the children in day care in the United States are cared for in family day care homes, and the demand for this type of care is expected to in- crease with the expanding labor force participation of young women. At the time this study was initiated in 1976, however, only a limited picture of family day care was available and that picture was based on a variety of small-scale studies of indeterminate generalizability and limited scope. This national study of family day care was designed to provide the first comprehensive detailed in- formation about family day care and thus establish a foundation for sound day care policies and programs. > The National Day Care Home Study attempted to describe the nature of family day care in many different types of settings, among different types of people and for children of different ages. It explored family day care along many different dimensions, ranging from community cultural patterns and the structure of the day care market to daily experiences of children and caregivers, parental preferences and satisfaction, costs and characteristics of the family day care population. The study thus provides the first glimpse of family day care through the multifaceted prism of the day care environment. Findings of the National Day Care Home Study (NDCHS) are presented in a seven-volume Final Report Series, of which this summary report, Family Day Care in the United States: Summary of Findings is the first volume. This volume summarizes information from the major research reports for the general reader who does not require the extensive detail presented in other reports. A description of each volume in the NDCHS Final Report Series is presented inside the front cover. Background Before turning to a description of the NDCHS, it will be useful to discuss the social and policy context of the study. The Demand for Family Day Care As a result of inflation, changing family structure and the career aspirations of young women, mothers are entering the work force in ever increasing numbers and, at the same time, the average age of children when their mothers enter the work force is decreasing. These fundamental changes in labor force participation have made day care an increasingly important social and economic support for families. In 1950, only 20 percent of all mothers with children under 18 were employed; by 1979, their labor force participation exceeded 50 percent. The largest increase during this period occurred among mothers of children less than six years of age, whose employment rate more than tripled from 14 per- cent to 45 percent. 1 Within the past decade the most striking increase has oc- curred for women with even younger children. For example, in the eight years from 1971 to 1979 the employment of women with children under three rose from 27 percent ot 40.9 percent. These increases in labor force participation of mothers are expected to continue through the end of the century, though at a slightly slower pace. By 1990, nearly half of all women with children under six will be working. For some segments of the population — for single parents, minority families and families with marginal incomes — employment rates already far exceed this figure. The substantial number of working mothers translates into large-scale de- mand for child care. In 1975 almost 30 percent of the country's 56 million families were using some form of day care. Approximately 7.5 million families regularly use care for 10 hours or more a week. Of this number, fully 45 per- cent use family day care. An additional 35 percent choose substitute care in their own homes, and 17 percent place their children in day care centers, nurseries, Head Start and other preschool programs (see Table l). 2 In terms of both the number of families using care and the number of children served, family day care constitutes the most widely used form of day care in the United States. In 1975, an estimated 1.3 million family day care homes served approximately 3.4 million full-time children (30 hours or more TABLE 1 .—Number of Children Aged 0-14 in Care 10 Hours a Week or More, 1975. Number of Children Proportion of Children in Care 3 Proportion of All Children** Family Day Care: Relative Nonrelative 2.8 million 2.3 million 25% 20% 06% 05% 5.1 million 45% 11% In-Home Care: Relative Nonrelative 2.2 million 1.8 million 19% 16% 05% 04% 4.0 million 35% 09% Preschool Programs: Day Care Centers Nursery School Head Start 0.8 million 1 .0 million 0.1 million 07% 09% 01% 02% 02% 00% 1 .9 million 17% 04% Other 0.4 million 03% 01% Total 1 1 .4 million 100% 25% a An additional 16.7 million children are in family day care less than 10 hours a week. b Total number of children = 47.9 million. per week), 2.8 million part-time children (10-29 hours per week) and 16.7 million children in occasional care (less than 10 hours per week). 3 More than half of the full-time children in family day care homes are under six years of age; the greatest proportion of these children are under three; and approx- imately 30 percent are aged three to five. Family day care also represents the most prevalent mode of care for the five million school children between 6 and 13 whose parents work. 4 The demand for day care, while already substantial, is expected to further increase as a result of the rising employment rates of mothers over the next two decades. This increasing demand coexists, paradoxically, with continuing declines in the total child population. The last 15 years have been marked by a steady decrease in the nation's fer- tility rate as well as in the absolute number of children under six (see Table 2). Although at the public school level this has led to a decline in enrollments, there is reason to believe that no such comparable decline has occurred in the level of day care for preschool children. The principal reason is that during this same 15-year period there has been a noteworthy increase in the labor force participation rates of women, especially for women with preschool children, as well as a shift in the patterns of care for their children. Whereas in 1965 only r about one-third of all preschool children of working mothers were cared for in institutional settings and by nonrelatives, by 1975 nearly two-thirds were cared for in these ways. Table 2. — Estimates and Projections of Number (in Thousands) of Children under Six Years with Mothers in the Labor Force, 1965-1990. 5 1965 1970 1975 1977 1980 1985 1990 Total Children 24,604 19,606 18,134 17,417 19,110 22,296 23,331 under Six Years. Children under 3,794 5,590 6,512 6,431 7,835 9,810 10,453 Six Years with Mothers in the Labor Force. Percentage of 15.0 28.5 35.9 37.6 41.0 44.0 44.8 Children under Six with Mothers in the Labor Force. As members of the post-war baby boom (1946-1964) reach maturity during the next decade, their sheer numbers will contribute to a marked increase in the number of preschool children and, consequently, in the demand for day care. Even using conservative estimates of fertility (2.1 children per woman), a 22 percent increase in the number of preschool children is projected over the com- ing decade (see Table 2). Since a continuing rise in the labor force participation of women with children under six can also be expected, an increasing demand for out-of-home day care can confidently be forecast. A closer look at growth in the rate of female labor force participation shows that increased employment for women with children under three is even more substantial than that for women with children under six. This signals an im- portant change in the composition of the demand for day care. Until recently, substitute care for children under three has been supplied primarily by members of the extended family, either in the child's own home or in the home of a relative. With the restricted availability of relative care due to such factors as a decline in the number of three-generation households, other sources of day care are likely to respond to this increased demand. The two most likely market responses are an increase in the number of infant slots in day care centers and an increase in the amount of family day care provided by nonrelatives. Since parents have traditionally shown a distinct preference for family day care over centers or other group facilities for infants and young toddlers, it is expected that the family day care market will absorb most of the increased demand for the care of children under three. L Structure of Family Day Care Family day care in the United States encompasses three distinct forms of care distinguished by their regulatory status and administrative structure. First, a large number of informal, unregulated family day care homes operate apart from any regulatory system or external administrative structure. A sec- ond group of licensed or registered homes exist within a regulatory framework but, except for minimal ties with the regulatory agency, operate independently in much the same way as unregulated homes. Finally, sponsored family day care homes are also regulated but operate as part of a day care system under .the administrative umbrella of a sponsoring agency (see Figure 1). / Historically, family day care was provided without charge by realtives or bartered between friends and neighbors in an informal exchange of services. I Informal, unregulated family day care, the oldest, most prevalent form of out- ^"6f-home care in this country, has long been an American tradition. Studies have consistently shown that informal arrangements are the overwhelming norm in family day care, accounting for over 90 percent of all family day care in the United States. Traditional, unregulated arrangements are estimated to include approximately 1.8 million homes serving nearly 5 million children for 10 hours a week or more (see Figure 1). Although the nature of informal family day care is changing with the gradual disappearance of the extended family and the declining availability of young mothers at home to care for their neighbors' children, these very per- sonal day care arrangements are often considered to reflect the cultural homogeneity of kinship and friendship networks in a given community. They are valued for the closeness of family values, life-styles and child-rearing pat- terns which provide stability and cultural continuity for young children. At the same time, informal family day care is seen as isolated from the resources available to the rest of the day care community and lacking visibility to parents outside the immediate neighborhood. Most parents now pay for their child care, and the choice of nonrelatives for the care of children under six is nearly as prevalent as relative care. Most arrangements, however, still involve friends, neighbors or acquaintances. Even when the caregiver is initially a stranger, she usually lives in the same neighborhood as her client families. As noted above, family day care in the United States overwhelmingly con- sists of informal, unregulated arrangements, which account for about 94 per- cent of all family day care. The remaining six percent of family day care homes are formally licensed by or registered with a regulatory agency. Regulated pro- viders are required to meet state and local day care requirements and, if serv- ing federally subsidized children, to meet federal day care requirements as well. The content and level of day care standards varies from state to state, although all cover such topics as group composition, health and physical safety, food and nutrition and caregiver qualifications. Within the broad category of regulated family day care are two subcategories: caregivers who operate independently and those who are affiliated with a sponsoring agency. Together these family day care homes account for approximately 6 percent of all family day care in the United States, encompassing an estimated 115,000 listed homes. Of these, approximately 74,000 are operating homes which serve 304,000 children (see Figure 1). The largest proportion of regulated providers operate independently like their unregulated counterparts, but in addition maintain formal ties to the regulatory agency. There are currently 85,000 licensed or registered family day care homes listed by state regulatory agencies. National Day Care Home Study data, however, suggest that only slightly more than half of such homes actually provide child care at any given time. Accordingly, the current estimates of regulated care assume approximately 44,000 operating family day care homes serving 175,000 children. The second major category of regulated family day care homes consists of providers who operate as part of day care systems or networks of homes under the sponsorship of an administrative agency. Accounting for approximately 2 percent of all family day care providers and 26 percent of regulated homes, these 30,000 day care homes serve approximately 129,000 children or 42 per- cent of all children in regulated family day care. This figure represents only 3 percent of all children currently in family day care (see Figure 1). Sponsored homes, in general, serve children whose care is subsidized, and often sponsored providers have access to a range of services, such as caregiver training and client referral, not available to other providers. Most of the children in family day care whose care is subsidized through Title XX of the Social Security Act are found in sponsored homes, as are all children who receive meals subsidized by the USDA Child Care Food Program. The trend toward organization of family day care homes into systems is a fairly recent development but one with important implications for future day care pro- grams and policies. Overview of the NDCHS Despite the widespread use of family day care and its importance as a fun- damental characteristic of contemporary American society, little has been known about the range of typical family day care environments, cultural pat- terns in caring for children, similarities and differences among unregulated, regulated and sponsored care or the dynamics of the family day care market. Similarly, little has been known about how to support families and caregivers in providing high quality day care in home settings. As mothers of young children increasingly enter the labor force and more children require substitute care at younger ages than ever before, there is a critical need for family day care which meets the diverse needs in this country at a cost which parents and taxpayers can afford. Needed support for the family day care community can be provided, in part, through development and implementation of sound standards for quality care, through training and technical assistance for providers, through improvement of service delivery systems and through assistance to parents in finding and cc o cc o LLI LU < ,_ OC 3 Q OC < o >- < Q >- CD CO 3 O > O 0) ~ E 5 o CO ^ Q ^r < w o c I sz a> 0>"0 (0 LL •D 2 a>oo CD £ Q-Cvj CO o o 3 CD on 03 *~ & m C Of^ ^ " (01 Q O « i: E co total) if Regul home Total Family 0>T3 C.C3 o o oo 1! tj- r^ re irf<- 03 LL C CD •D C CD QL CD ■D C children (3% of (58% c listed homes operating homes e = 4.0 children per O OOO) O O IT) 03 irf or-. q5 co r- mco > r- 00 ^ 03 CD zz CO O total) home CD rx 2* O m co Q ^ -coE2 OJ OT3 E co LL ■D hildren sted horn erating h = 4.1 chil £ 0= Q. M 3 OO ° OOO O) oo in to O) "^irr 1 ^. q3 CD CD DC O t-C0 > co y- r- 03 CO O >> TJ CO Q) Q 3 oi >« otal) Reg ome: ome I CO r "s-c-c LL (3% o (42% erating ren per "D CD O CO id op child c o £ So a 2 T3-* (/) fi ~ II O (/) " T3 £ O — 0) 8 8? co CT>* °. Q> c\j o > zz r- CO 03 O) CD rx maintaining child care which meets their particular family needs. The National Day Care Home Study was initiated to provide a comprehensive base of in- formation to enhance the development of these important areas and to pro- mote increased effectiveness and efficiency in home-based care. The National Day Care Home Study represents the first national study of family day care and the first attempt to describe the ecology of family day care as a complex social system. It is the first major study to examine all of the prin- cipal family day care participants— the caregiver, the children in care, their parents, day care program administrators and the community institutions which complete the day care milieu. All major forms of family day care are represented in the National Day Care Home Study, including regulated homes, sponsored homes and the first large sample of informal, unregulated family day care homes ever studied. In addition, the National Day Care Home Study represents the only study of national scope to systematically observe the care of children in home en- vironments using sophisticated and carefully tested instruments. Finally, the study focuses on understanding cultural diversity in family day care among the three groups who together constitute the largest users of family day care: (non- Hispanic) Whites, (non-Hispanic) Blacks and Hispanics. Major objectives of the NDCHS were to: • describe cultural and demographic patterns of family day care; • describe the range of program elements, services and administrative structures in family day care homes; • describe the nature of care provided and document the»day-to-day ex- periences of caregivers and children; • identify similarities and differences among unregulated, regulated and sponsored homes; • explore parents' needs, preferences and satisfaction with their day care arrangements; • describe the community context for family day care and identify major factors affecting availability and utilization; and • identify major economic factors and document the costs of family day care — to the parent, to the government and to the provider. These objectives reflect the principal goals of the National Day Care Home Study: to develop a comprehensive and detailed description of family day care and to provide useful information for the improvement of day care quality, formulation of sound day care policies and assistance to the day care com- munity. Spanning a four-year period from 1976 to 1980, the National Day Care Home Study was completed in four stages. Phase I of the study entailed development of the research design, instrumentation and operational plans. In Phase II the study was implemented in Los Angeles, the first of three sites; this phase constituted a large-scale pilot test of all design elements and field pro- cedures. During Phase III, the study was extended to Philadelphia and San Antonio. Data from all three communities were analyzed and findings reported in Phase IV. 8 The study sample consisted of caregivers, parents, children, day care pro- gram directors, agency officials and others who were involved with family day care. Initial interviews were held with all individuals identified as caring for one or more children other than their own. A smaller sample was later selected from this group for more detailed study. Interviews were also conducted among the parents of children in care, with agency officials, day care systems directors and other individuals in the community. In addition, direct observa- tions of children and their caregivers were conducted in the family day care homes. Overall responsibility for management and technical direction of the National Day Care Home Study rests with the Day Care Division of the Ad- ministration for Children, Youth and Families in Washington, D. C. Four research organizations participated in this study. Development of the research design, field management procedures and interview instruments during Phase I was carried out by a research consortium composed of Westat, Inc. of Rockville, Maryland; Abt Associates Inc. (AAI) of Cambridge, Massachusetts; and the Center for Systems and Program Development (CSPD) of Washington, D.C. Caregiver and child observation systems were developed by SRI International of Menlo Park, California. Beginning with Phase II, AAI became Research Contractor for the study with CSPD as sub- contractor; SRI remained as Observation Contractor. In addition to the agency and research organizations conducting the National Day Care Home Study, a national consultant panel was established during Phase I to provide formative advice, consultation and peer review. The consultant panel, representing a range of relevant specialties, participated in the study design, implementation, analysis and reporting of results. The panel included Black, White and Hispanic consultants to ensure sensitivity to issues of concern to the populations most frequently served by family day care. Early in the study, the minority group members of the panel formed a Minority Task Force to identify technical and policy issues of particular significance for minorities and to offer broad procedural guidelines for addressing these con- cerns. Notes 1. Congressional Budget Office, Childcare and Preschool: Options for Federal Support (Washington, D.C: U.S. Government Printing Office, 1978). 2. Unco, Inc., National Childcare Consumer Study: 1978, prepared for the Department of HEW, Office of Child Development (1975), Volume I, "Current Patterns of Childcare use in the United States." 3. Ibid. 4. Ibid. 5. Hofferth, Sandra L., "Day Care in the Next Decade: 1980-1990," Journal of Marriage and the Family, August 1979, p. 650. Chapter 2 DESIGN AND METHODOLOGY* Underlying the principal objectives of the NDCHS were several basic assumptions about the structure and dynamics of family day care which together molded the framework for the study design. First was the belief that family day care, as a complex social phenomenon, is intricately woven into the fabric of American society, as are American families themselves. In order to portray this cultural richness and diversity, it was necessary to develop a research design which included all of the major populations in family day care. This requirement had far-reaching consequences for all elements of the design and was an essential consideration in the selection of the three study sites, the three ethnic groups and the wide range of child ages represented in the study. In addition, this factor necessitated the inclusion of the broadest possible spec- trum of social and economic characteristics, including families eligible for sub- sidized care as well as those from more affluent segments of society. Likewise, the traditional role of the extended family in informal care arrangements re- quired that both relative and nonrelative care be represented. Accordingly, the National Day Care Home Study was designed to allow for both extensive sam- ple stratif action and natural variation along a large number of dimensions. A second assumption which ultimately focused the study design was con- cerned with the structural complexity in family day care, which is simultaneously a unique set of personal relationships, a traditional community service and a part of the broader day care picture, involving federal, state and local governments, child advocacy groups, various child-related programs and other community institutions. This assumption led to the inclusion of all ma- jor family day care participants: the providers, children and their parents, day care program administrators, agency officials and community child advocates. The decision to include a wide range of participants reflected an ecological ap- proach to the study, which attempts to describe family day care from each of the various perspectives as well as to arrive at a picture of family day care as a whole. A third supposition was that the structure and function of family day care would be heavily influenced by social and economic forces both in the larger day care community, (e.g., federal subsidy and regulation) and in the more im- mediate community environment (e.g., local day care initiatives, state •Greater detail on the NDCHs design and methodology is presented in the NDCHS Final Report, Volume II, Research Report, prepared by Abt Associates and in the volumes from each component. 11 regulatory practices, funding priorities and the availability of such resources as public information, training, referral). The assumption required that the study be conducted in communities that reflected a range of regulatory practices, in- cluded both subsidized and nonsubsidized care and exhibited a range of local family day care conditions. Another consideration was the fact that both federal and state governments are extensively involved in family day care — as purchasers of care, as regulators of quality standards, as providers of support services, and as general advocates for children. Yet little has been known about the effect of such involvement or how best to support family day care. Because so little in- formation was available and the range of important issues was extremely broad, a study yielding basic data on a number of dimensions was considered preferable to a more narrowly focused one. Finally, the state-of-the-art with regard to measurement of child behavior and child development affected the study's design. Because of the increasing concern among parents, educators and developmental psychologists over the questionable validity of standard measures and approaches, particularly for children under three and for minority populations, the NDCHS focused on the day-to-day experiences of children in their day care environments rather than on developmental outcomes. This decision required a heavy emphasis on natural observations of both caregivers and children. Because existing obser- vation instruments were inadequate to meet study needs, a large-scale effort was undertaken to develop and test a sophisticated observation system for documenting caregiver and child behaviors and for characterizing the social climate of the home. As a consequence of these considerations, the NDCHS research design was constructed to provide systematic variation in type of community, regulatory environment and geographic location across the three urban communities selected for the study. Variation was also sought in ethnicity of caregivers (Blacks, Whites and Hispanics) and regulatory status of homes (sponsored, regulated, unregulated). This design is presented in Table 3. Table 3. — National Day Care Home Study Design: Ethinicity and Regulatory Status. Ethnicity of Regulatory Status 3 Caregiver Sponsored Regulated Unregulated White Sponsored home/ Regulated home/ Unregulated home/ White caregiver White caregiver White caregiver Black Sponsored home/ Regulated home/ Unregulated home/ Black caregiver Black caregiver Black caregiver Hispanic Sponsored home/ Regulated home/ Unregulated home/ Hispanic caregiver Hispanic caregiver Hispanic caregiver *This basic nine-cell design will appear repeatedly as findings are presented in tables throughout this report; the headings / ihmr a y of ( ureffiver and Regulatory Status will be omitted in subsequent tables in order to streamline the presentation of data. 12 The Study Sites The National Day Care Home Study was conducted in three urban sites — Los Angeles, Philadelphia and San Antonio — selected for geographic, socioeconomic and ethnic diversity as well as for diversity in their approach to the regulation of family day care and in the types of family day care homes available for the study. The choice of three communities represents an attempt to balance the need for generalizable information with that for a detailed and comprehensive understanding of family day care. The first requirement called for the selection of a sufficient number of sites to estimate both variability and commonality along many dimensions of family day care as practiced in dif- ferent types of communities. The second criterion demanded a small enough number of sites to mount an intensive, well-controlled investigation of all ma- jor elements. Three sites were considered the minimum necessary to address the first objective and the maximum feasible for achievement of the second. Statistical power analyses also indicated that nine study sites— far more than study resources could bear — would be required for a significant increase in generalizability of findings beyond that possible with three sites. For these reasons, careful selection of three communities was considered optimal for the NDCHS. Site selection was further limited to urban settings, both as a result of the restriction in number of sites and because of other technical design re- quirements. For purposes of comparability, a rather large number of unregulated, regulated and sponsored homes in each community was re- quired — as were Black, White and Hispanic homes and homes serving children of various ages. No rural or semiurban sites were located which met these criteria. In addition, the selection of specific sites was heavily dependent on the total range of characteristics represented by the three communities in combination. For example, it was necessary to have both geographic dispersion and wide cultural latitude as well as a range of community day care usage patterns. Similarly, various regulatory approaches were needed, as was an adequate representation of public and private, formal and informal care. Likewise, a fair representation of care provided by relatives and nonrelatives, paid and nonpaid care, various programmatic orientations and other facets of day care were needed to adequately characterize family day care. Because subsidized care is one of the principal areas of impact for federal day care policy, the study sites were required to contain a substantial number of families who were currently and potentially eligible for family day care sub- sidies under various federal and state programs; likewise, an adequate number of family day care homes serving this population had to be included. To study other areas of policy significance such as training, technical assistance and consumer support, the sites in combination were selected to reflect differences in degree of active day care involvement on the part of the day care commu- nity. 13 A final site selection criterion required acceptance of the study on the part of community and day care leaders. Because of the extensiveness of field opera- tions, the large amount of data to be collected and the difficulty of identifying family day care homes, active community support of the study represented one of the most critical factors for the success of the research. In summary, the three NDCHS sites were selected according to the follow- ing criteria: • geographic diversity; • socioeconomic diversity; • cultural diversity; • diversity in regulatory approach to family day care; • a large total amount of family day care; • adequate numbers of regulated, unregulated and sponsored homes; • a large pool of parents and caregivers eligible for day care subsidies; and • acceptance of the study by the community. Los Angeles, Philadelphia and San Antonio were among the few U.S. cities to meet all criteria; these three cities represent the optimal mix of sites taken as a whole.* Within-Site Study Design The National Day Care Home Study included several distinct samples, representing caregivers, children in care, parents and individuals from the wider day care community. Core Samples Because of the need for both broad profiles and detailed information, a two- tiered sampling strategy was employed. In the first stage core samples were ob- tained. First, a core sample was constructed of all providers identified during an intensive site development effort — a total of 793 caregivers. This sample was composed of all respondents who cared for children other than their own, regardless of ethnicity, regulatory status, number or ages of children in care, the total amount of care provided or any other factor. A preliminary interview with caregivers in this core sample formed the basis for many of the summative profiles of family day care populations and home characteristics presented in this volume. The core caregiver sample also provided the foundation for all of the subsequent samples of caregivers, children and parents selected for in- depth study as the research progressed. The distribution of the core caregiver sample is given in Table 4. The un- equal distribution of sponsored, regulated and unregulated caregivers results from deliberate oversampling in the unregulated and regulated sample cells. This sampling strategy was used in anticipation of higher refusal and attrition rates among caregivers not linked with an active day care system, particularly among providers who operated informally outside the regulatory structure. •It should be remembered that because data obtained in the NDCHS are limited to three sites, findings do not reflect a picture which is statistically representative of the entire country. 14 Table 4. — Distribution of Core Caregiver Sample. Sponsored Regulated Unregulated Total LOS ANGELES White Black Hispanic 18 20 24 71 40 22 49 49 36 138 80 82 Total . . 62 133 105 300 SAN ANTONIO White Black Hispanic 9 12 61 13 31 57 48 80 118 70 123 Total .. 21 105 185 311 PHILADELPHIA White Black Hispanic 23 37 21 39 29 33 73 109 Total .. 60 60 62 182 ALL SITES White Black Hispanic 41 66 36 150 92 53 135 101 116 367 259 205 ( 46.2%) ( 32.7%) ( 25.9%) Total .. 143 (18.0%) 298 (37.6%) 352 (45.0%) 793 (100.0%) The core child sample consisted of all children identified in the homes of the core sample of caregivers, including both day care children and the caregiver's own children who were at home during the day. This core child sample formed the basis for general profiles of the child populations. Parents of children in care were not sampled at this stage of the study. Main Study Samples The second part of the two-tiered sampling strategy involved the selection of four samples for the main study. These main study samples were derived from the core samples just described. They included a caregiver interview sample, an observation sample of both caregivers and children, a parent interview sample and a sample of other actors in the day care community. The main study caregiver interview sample was stratified by ethnicity and regulatory status. The former was selected as a stratifying variable to assure adequate representation of Black and Hispanic caregivers in the sample. If the caregiver sample had, instead, been constructed in proportion to the ethnic composition of the population, the resulting sample would have been com- posed primarily of Whites and, as will be shown below, most of the richness and diversity of family day care would have been missed. Regulatory status of the home was selected as the second stratifying variable because of the policy 15 interest in drawing separate and distinct inferences for sponsored, regulated and unregulated homes. Statistical considerations led to a goal of 16 caregivers for each of the nine cells, in each of the study sites. This was to produce an overall caregiver sam- ple size adequate to draw inferences both within sites and across sites about the differences and similarities across caregivers. Some exceptions to the rule of 16 proved necessary, however, because few sites in the country could match the ideal that had been set. There were few Hispanic caregivers in Philadelphia, and thus, none were selected in Philadelphia for the study design. There were no White sponsored caregivers in San Antonio, and few Black or Hispanic sponsored caregivers. The study design was adjusted to meet these realities. The planned distribution of caregiver interviews for the main study, taking these changes into account, is presented in Table 5. As Table 5 shows, the planned distribution called for a total of 352 interviews. In fact, 501 caregivers were actually selected for the main study; some oversampling was undertaken to ensure adequate sample sizes, especially among unregulated caregivers where attrition was expected to be higher. Caregivers were selected for the main study according to a number of criteria. They were required to care for at least one child aged 12 to 60 months for at least 20 hours a week and to provide care for pay. These selection Table 5 . — Planned Distribution of Caregiver Interview Sample for the Main Study. Sponsored Regulated Unregulated Total LOS ANGELES White Black Hispanic 16 16 16 16 16 16 16 16 16 48 48 48 Total . . 48 48 48 144 SAN ANTONIO White Black Hispanic 8 8 16 16 16 16 16 16 32 40 40 Total . . 16 48 48 112 PHILADELPHIA White Black Hispanic 16 16 16 16 16 16 48 48 Total . . 32 32 32 96 ALL SITES White Black Hispanic 32 40 24 48 48 32 48 48 48 128 136 88 ( 36.4%) ( 38.6%) ( 25.0%) Total .. 96 (27.3%) 128 (36.4%) 128 (36.4%) 352 (100.0%) 16 criteria were necessary to eliminate incidental day care arrangements. Although these incidental arrangements are important to parents, the principal concern in this study was with day care of a more substantial nature. These criteria further reflect the presumption that incidential arrangements would be less significant for the children involved. The caregivers selected for the main study participated in follow-up interviews and observations that provided most of the detailed information related to caregiver characteristics, the range of services provided by caregivers, day care costs, and the observed characteristics of day-to-day activities among caregivers and children in family day care homes. From the core caregiver sample in Table 4, a subsample of 305 homes was selected for in-home observations of both caregivers and children. The obser- vation sample consisted of children in the NDCHS main study homes who were between 12 and 60 months of age, who were in care 20 hours a week or more and who had parental permission to participate in the observations. The distribution of the sample is given in Table 6. This sample was divided between children aged 12 to 35 months and children aged 36 to 60 months to assure ade- quate representation of these two developmentally distinct stages. Whenever possible, two children were observed in each home — one child in each of the Table 6.— Distribution of Observation Sample. [Main Study] Sponsored Regulated Unregulated Total LOS ANGELES White Black Hispanic 14 ♦11 16 15 13 7 13 2 8 42 26 31 Total .. 41 35 23 99 SAN ANTONIO White Black Hispanic 6 11 24 7 19 19 14 17 43 27 47 Total .. 17 50 50 117 PHILADELPHIA White Black Hispanic 15 18 12 17 14 13 41 48 Total .. 33 29 27 89 ALL SITES White Black Hispanic 29 35 27 49 37 26 48 29 25 126 101 78 ( 41.3%) ( 33.1%) ( 25.6%) Total . . 91 (29.0%) 112 (36.0%) 128 (32.8%) 305 (100.0%) 17 two age categories. Infants under 12 months of age and school-aged children were included in the observations whenever they interacted either with the caregiver or with focus children aged 12 to 60 months. The parent interview sample was selected from the 501 family day care homes included in the main study. One parent was selected from each of the NDCHS study homes. Only parents who had a child aged 12 to 60 months in care for 20 hours a week or more and who paid a fee for care were eligible. Parent interviews were conducted with a total of 348 families in the three sites — 105 in Los Angeles, 122 in San Antonio and 121 in Philadelphia (see Table 7). Data from this study component relate to such questions as parent needs, preferences and satisfaction with their family day care arrangements. Table 7.— Distribution of Parent Sample. [Main Study] Sponsored Regulated Unregulated Total LOS ANGELES White Black Hispanic 12 11 12 15 11 12 14 7 11 41 29 35 Total .. 35 38 32 105 SAN ANTONIO White Black Hispanic 5 10 25 5 20 24 17 16 49 27 46 Total .. 15 50 57 122 PHILADELPHIA White Black Hispanic 24 28 14 24 18 13 56 65 Total . . 52 38 31 121 ALL SITES White Black Hispanic 36 44 22 54 40 32 48 29 25 146 121 81 ( 42.0%) ( 34.8%) ( 23.3%) Total . . 91 (29.0%) 126 (36.2%) 120 (34.5%) 348 (100.0%) In addition to the caregiver, child and parent samples, NDCHS staff inter- viewed a sampling of other individuals who were involved with family day care in each community. This sample included representatives of state and federal funding and regulatory agencies, day care program administrators, members of day care and child advocacy organizations and community leaders. Altogether, over 30 such individuals were interviewed in each site. Information from this component of the study formed the basis for site case studies, which describe the geographic, social and political context for family day care in each 18 of the three study communities. Finally, administrative staff from family day care sponsoring agencies were extensively interviewed to document the opera- tions, programs, services and costs of providing family day care through day care systems. Field Implementation Because of the complexity of the National Day Care Home Study and the lack of previous large-scale family day care research to guide the implementa- tion process, a phased implementation strategy was employed. Phase I, a developmental stage, was designed to obtain a preliminary notion of impor- tant variables, test sampling and field management procedures, assess the usefulness of interview items and observation codes, and develop detailed specifications for field operations, contractor coordination and data analysis. During Phase II the study was conducted in the first site (Los Angeles) as a large-scale pilot study. All aspects of the study were fully tested and later re- fined as a result of the pilot experience. The study was extended to the two re- maining sites (Philadelphia and San Antonio) during Phase III. Preliminary data from Los Angeles were used to finalize data analysis plans for the study as a whole, and the complete three-site analysis was conducted during Phase IV. This phased strategy proved cost-effective for the NDCHS because it minimized the risk in mounting a complex and difficult field effort, while simultaneously ensuring increased efficiency in subsequent field operations. As a result of the successful pilot experience, only minor modifications were required in data collection instruments and procedures. These refinements facilitiated subsequent field operations and enhanced the quality of data. In addition, this strategy allowed for a longitudinal study in Los Angeles at very little extra cost relative to the substantial information gain. Finally, the op- portunity to conduct extensive reliability studies on the newly developed obser- vation systems as the study progressed resulted in a demonstrably high quality of data for this unique and technically difficult component of the study. Recruitment of Family Day Care Providers The first major implementation effort of the NDCHS was a 25-city survey of family day care providers conducted during Phase I. The purpose of this survey was to determine the feasibility of identifying family day care homes through standard household probability sampling techniques and to obtain preliminary profiles to guide other design decisions. This survey demonstrated the difficulty of finding family day care homes using a household probability sample and established the necessity of mounting exhaustive site development efforts with extensive on-site preparation and multiple recruitment techniques. In particular, a different approach was needed to identify sponsored and regulated homes, for which lists were available, than was needed to identify unregulated, unlisted caregivers. 19 To assure the comparability of sponsored, regulated and unregulated homes identified in this study, several target areas were selected within each study site and study activities restricted to these areas. Comparisons could more readily be made within such communities because of their greater sociocultural homogeneity. Many of the target areas in each site were selected by identifying areas with the greatest concentration of sponsored and regulated homes in the site. (It was presumed that any communities with regulated care would have unregulated care as well; this assumption was borne out during the study). To supplement these target areas we sought out additional areas which had a preponderance of families whose children would be eligible for day care sub- sidies. For sponsored care, it was a relatively straightforward process to identify homes and obtain the cooperation of caregivers. Not only were lists of these homes available from family day care systems and regulatory agencies, but providers were also encouraged by local agencies to cooperate with the study. Most family day care systems briefed affiliated caregivers about study objec- tives and encouraged their participation. This group of sponsored homes, as a result, had almost 100 percent participation. The identification of regulated homes was more difficult because the lists obtained from day care regulatory agencies were found to be out-of-date. Of the 766 regulated providers randomly selected from these lists for the study, only 390 (51%) were in fact operating family day care homes. In each of the study sites, about one-quarter of the caregivers listed were no longer providing care and another quarter could not be reached — either because they had moved away without leaving a forwarding address (7.8%) or because they were repeatedly found not to be at home (14.8%). Three-quarters of the regulated caregivers contacted during the course of the study expressed a will- ingness to be interviewed. Of those who refused, some did so because of the amount of time it would have involved. Others simply indicated that they did not want to be bothered or seemed fearful of being reported to local authorities if found out of compliance with regulations. Identification of informal, unregulated family day care presented the greatest challenge; although these homes provide the bulk of family day care, they are not easily found within the community, even by parents looking for care, unless they are part of the kinship and friendship networks of the neighborhood. Unlike day care centers, which are relatively visible to the public, family day care homes are not immediately distinguishable from other residences. Furthermore, unregulated caregivers, once identified, were less in- clined to participate than were providers in either of the other two groups. Many, for example, did not consider themselves part of the day care commu- nity. This was particularly true of providers who cared for related children. Some considered interviews a violation of their privacy and objected to strangers entering their homes (both common problems in household survey research). Still others expressed concern about operating in violation of local or state laws, and some were themselves undocumented persons working il- 20 legally in this country (a common occurrence in the Hispanic communities of both Los Angeles and San Antonio). These barriers, while pronounced in the unregulated sample, proved surmountable by intensive site development and involvement of community members in the research. An extensive public relations campaign launched the identification of unregulated care in each community. The aim was to disseminate information about the study, to gain the cooperation of family day care providers and to give community residents an understanding of how and why the study was be- ing conducted. Brochures familiarized caregivers with study objectives, described what participation in the study would involve and requested help in locating other family day care homes. Letters from licensing and registration officials accompanied the brochures, helping to promote the study's legitimacy. Hundreds of posters explaining the study and asking for help were also hung in locations likely to catch the eye of local residents, especially parents with young children. In Philadelphia and San Antonio, media coverage was excellent; it included TV spots, radio and television interviews and newspaper articles. To facilitate the identification of unregulated care, field staff were recruited who had extensive knowledge of target areas. Interviewers from "down the block," who knew the community leaders and were familiar with local agen- cies, were more likely to gain the trust of caregivers, especially those operating unregulated homes. This neighborhood approach worked well in Philadelphia and San Antonio but was considerably less successful in Los Angeles. It had been assumed that Los Angeles, like most American cities, was divided into small geographic areas with identifiable socioeconomic and cultural characteristics — neighborhoods with which residents identify or have strong ties. Initial staff experiences, however, belied these expectations. It was ex- tremely difficult to identify neighborhood leaders in Los Angeles with any knowledge of informal day care operating in their communities. This absence of traditional neighborhoods required a less systematic and more time- consuming approach to the identification of unregulated care at that site. In general, the stategies employed by site development staff in locating unregulated family day care depended on their knowlege of target com- munities. Community leaders and social service agencies were briefed about the study and asked for their support. Day care centers and elementary schools were visited to identify homes serving children on the centers' waiting lists and to identify homes providing after-school care. Contact was also made with welfare offices, YWCA's, CETA, unemployment offices, local universities, women's centers and other community agencies that might have a knowledge of family day care. Over one-third of the unregulated homes in Philadelphia and San Antonio were identified through these sources, but because of the community structure in Los Angeles, few homes there were identified with these techniques. In selected study neighborhoods, door-to-door canvassing was employed in an attempt to locate unregulated care. This approach is excellent from a 21 technical standpoint but was feasible in only a few areas because it is expensive and requires intimate familiarity with the neighborhood on the part of the in- terviewers to obtain the caregiver's trust and cooperation. In places where direct door-to-door canvassing was used, very good results were obtained. Word-of-mouth and a more informal canvassing approach also proved suc- cessful in identifying homes. Field staff spoke with local mail carriers and ap- proached residents and managers of apartment buildings. Others drove through study neighborhoods and located homes with playground equipment or groups of children in the yard. This was particularly effective in low-income areas of San Antonio, where cramped housing quarters and oppressive sum- mer heat made it necessary for children to spend much of their time outdoors. Another effective recruitment approach was to ask family day care pro- viders if they knew of other caregivers who might wish to participate in the study. Caregivers were somewhat reluctant at first to recommend other homes. Only one-quarter (27%) of the caregivers in Philadelphia and San Antonio in- itially indicated that they knew others who provided care. However, once a relationship of trust had been established, twice as many caregivers referred the interviewer to other family day care homes (55%). Field staff also talked with residents in libraries, clinics, recreation centers and shopping areas. In addition, bulletin boards and newspaper ads were scanned in the search for homes. However, only a small number of leads were identified from newspaper ads, primarily because city-wide papers in Los Angeles and Philadelphia restrict advertising to regulated homes.* Representativeness of the Sample The samples of sponsored and regulated homes appear to be excellent. Because there were lists of homes from which to randomly sample and because of the consistently high level of cooperation among sponsored and regulated providers, NDCHS homes appear to be representative of these types of family day care homes. For unregulated homes, it is difficult to scientifically verify the quality of the sample obtained. The unregulated homes identified represent approximately 30 percent of all unregulated care estimated by officials to exist in Philadelphia and almost half of that in San Antonio. Although a substantial proportion of existing unregulated family day care appears to have been in- cluded in the study, this alone does not provide evidence of the sample's representativeness. In fact, many unregulated providers felt uncomfortable about participating in the study, and thus, the sample probably under- represents the more isolated segment of this population — though many such providers were persuaded, in time, to participate. The study also underrepresents unregulated part-time family day care homes — those caring for children for fewer than 15 hours per week. The Na- Although similar restrictions currently do not exist in San Antonio, few homes were identified in that site in this way. Caregivers clearly use other methods for obtaining children to care for in their homes. 22 tional Child Care Consumer Study conducted in 1975 demonstrates the prevalence of such informal part-day arrangements. The paucity of such homes in the NDCHS core sample seems to reflect the fact that women caring for children for only a few hours per week tend not to identify themselves as providing child care. This underrepresentation was not a concern, however, because the NDCHS focused on full-time care, the kind that would be of most use to a working parent. In fact, for the major portion of this study, only caregivers providing care for 20 or more hours per week were included. Overall, based upon intimate knowlege of the communities in the study and extensive experience in the field, NDCHS staff believe that the sample of unregulated homes obtained is an excellent representation of all unregulated caregivers who care for children for more than 20 hours per week. 23 : ng nutrition trition in a gula- ♦es Chapter 3 STATE REGULATION OF FAMILY DAY CARE* The regulatory environment of family day care in this country is composed of a melange of federal, state and local regulations often inconsistent in their content and application. Within each state, however, the picture is somewhat simplified. Federal regulations apply mainly to day care purchased under Title XX of the Social Security Act, and this constitutes a relatively small propor- tion of family day care. Thus state and local regulation has the greatest impact on the local context. Since 1940 one of the responses of states to the increased use of day care has been the promulgation of standards for the care of children in day care centers and in family and group day care homes. Prior to 1940, only one state had standards specifically referring to day care and that state regulated only day care centers. 1 By 1957, 41 states regulated group day care and 14 states regulated family day care homes. The primary method of regulation was licensing, and by 1968 day care licensing was "operating to some extent or as a very well established service in all but one of our 54 jurisdictions (including Washington D.C. and the Territories)." 2 A 1971 Office of Child Develop- ment** report on day care licensing found that family day care homes were regulated in 48 states: 39 of the states made licensing mandatory, three made it voluntary, and six states certified family day care homes. 3 States usually define a family day care home as a private home in which regular care is provided for 6 or fewer children, including the caregiver's own, for any part of a 24-hour day. There are, of course, variations in state regula- tions; some states place a ceiling on capacity at four and others permit larger numbers if the children are siblings. States and locales use a variety of regulatory mechanisms, the chief among which are licensing and registration. As indicated above, the majority of states license family day care homes. Licensure constitutes formal permission by the designated state or local agency to an individual to provide child care. The state agency has authority to set standards, make inspections to be sure the standards are met, set procedures for revoking a license and establish appeal mechanisms. Generally, a license is issued and renewed as long as all of the ap- plicable regulations are met. 4 ♦Information in this chapter is condensed from the NDCHS Final Report, Volume V, Family Day Care Systems Report, prepared by Abt Associates. **Now the Administration for Children Youth and Families, Washington, D.C. 25 Registration is a form of regulation which stresses caregiver self-inspection and consumer awareness. Under this system, the caregiver herself is respon- sible for determining whether or not she meets the applicable standards and for securing the necessary health and safety inspections. Instructions and pro- cedures for this self-review are typically mailed to the caregiver by the register- ing agency. Upon receipt of the completed information package by the agency, the family day care home is considered registered and has permission to operate. Inspections are not usually made unless a complaint is lodged. In some cases parents are provided a copy of the regulations, but the process dif- fers somewhat from state to state. State standards for the maximum number of children per family day care home are generally consistent with federal regulations. In 1970, 29 percent of the states allowed five or fewer children per home (including the provider's children), and 58 percent of the states permitted a maximum of 6 children. By 1976, 38 percent of the states allowed five or fewer children per home and 41 percent allowed a maximum of six children. Nearly all states (85%) allow no more than two children under two years of age. The total number of children allowed generally varies according to ages of children present with lower group size required for homes serving infants. States continue to set standards for group composition, health and physical safety, nutrition, caregiver qualifications and other aspects of family day care. Table 8 illustrates the changes from 1971 to 1976 in the number of states hav- ing standards for selected elements in family day care programs. The number of states requiring medical examinations, immunizations, and daily illness screening has increased slightly between 1971 and 1976. In 1976, 70 percent of the states required child medical exams, about half required immunizations and about one-third required daily illness screening. States are moving away from specific standards for facilities and appear to be relying more heavily on compliance with local codes to maintain safety and sanitation standards for Table 8.— Proportion of States with Standards for Certain Aspects of Family Day Care. Child Eligibility: Medical Exam Immunization Daily Illness Screening Nutritkrtf ". 7rrrT7T7>^ v . /Staff Qualification and Tr aining^_ ^J> Records ». Facilities: Space Health/Safety Code Compliance Program Equipment Staff/Child Ratio "Based on available data for 37 states. 26 Percent of States 3 1971 1976 64 70 37 51 27 37 90 64 92 70 80 81 88 37 76 67 68 81 68 86 90 97 family day care homes. States have also moved away from requiring nutrition standards for homes; 90 percent of the states had a standard for nutrition in 1971 compared to only 65 percent in 1976. Although the 1980 federal regula- tions do not require parent participation for family day care homes, ten states currently require that there be some observation, orientation or conferences. Most of the 50 states have laws governing the regulation of family day care homes. However, the sheer numbers of family day care homes and shortage of manpower caused by personnel ceilings in many states have precluded any comprehensive attempt at enforcement. By its very nature family day care is very costly for the states to supervise. A typical licensed home may have only three children. On a per-child basis the cost of licensing and monitoring a home is therefore burdensome in comparison with the costs of monitoring and licensing a day care center, where the average enrollment may be 50 or more. As a consequence, some states concern themselves officially only with homes receiving Title XX funds or other government monies, although most license only those caregivers who initiate a request, and still others have en- couraged the growth of family day care systems to facilitate funding and ad- ministrative functions. By dealing directly only with the system and not with individual homes, the state is able to shift much of the management burden from state to system staff, thereby enabling the state to handle larger numbers of subsidized homes. Alternatively, states have adopted the registration method as a less costly procedure per home for bringing family day care into a regulated environ- ment. While some critics cite registration as a regulating mechanism that does not provide adequate quality control, proponents of this self-certification method consider it preferable to bring larger numbers of providers into com- pliance through registration than to license only a small portion of providers at higher costs while backlogs accumulate and large numbers of providers remain unlicensed and uninformed about standards. How the family day care home is influenced or assisted by the regulatory agency is affected by a number of factors beyond the regulations themselves. Indeed, there is as much variation in regulatory activities across states which employ licensing as there are differences between licensing and registration. In a survey conducted by the Children's Foundation in 1978, six states were found to provide no regulatory support to providers, while 24 states provided minimal support in the form of 3-4 hours of pre-licensing training and materials available at state offices. Nineteen states provided extensive services including on-going training, technical assistance, workshops and seminars, monthly periodicals and toy lending libraries. 5 States which license family day care homes present a broad spectrum of dif- ferent licensing activities and services. How licensing functions and what an- cillary services are provided is often a function both of available funds and where the licensing units are lodged in state governments. In states such as Rhode Island, where the day care licensing unit is the same unit which ad- ministers Title XX funds, family day care providers are offered services such 27 as training — services made possible through the dual function of the unit. To the extent that such an office is also a sponsor for family day care homes for the Child Care Food Program, additional services are provided directly to the homes. In those states where licensing units are separate from Title XX ad- ministrative units, there is less likely to be a service package offered. San Antonio Regulatory Context In contrast to the other two study sites (Los Angeles and Philadelphia) San Antonio utilizes a registration system. Licensing of homes was abolished in 1975 through legislation which established registration. Family day care registration is handled within the Texas Department of Human Resources which also administers social services and child welfare. The Day Care Licens- ing Division is responsible for the regulation for all child care facilities. In general, the state's human services system is locally administered and super- vised by state agencies, a model which could be described as "laissez faire," allowing counties to respond broadly or stringently to locally perceived social needs. Over the last four years, Texas has used an average of 17 percent of its Title XX allocation for child care, but less than 5 percent of subsidized care in Texas is purchased from registered family day care providers. Conversely, Texas ranks as the state with the largest amount of center-based child care in the na- tion. Available data on the number and type of facilities indicate that there are approximately 1600 licensed commercial centers, 1000 licensed nonprofit centers, 150 licensed group day care homes and some 4000 registered family day care homes in Texas. Unlike Los Angeles and Philadelphia, where there are many organized family day care systems, San Antonio has only one. In- deed, over 95 percent of the registered family day care in this city is informal and unaffiliated in any way with the formal day care community. Regulatory Practices The Child Care Licensing Act was passed by the Texas Legislature in 1975. This act established the registration system, under which a registered family home was defined as: ... a child care facility which regularly provides care in the caretaker's own residence for not more than six children under 14 years of age, excluding the caretaker's own children, and which provides care after school hours for not more than six additional elementary school siblings of the other children given care, provid- ed that the total number of children including the caretaker's own does not exceed 12 at any given time. Officially DHR states that the purpose of registration is to "increase the ef- ficiency and effectiveness of the regulation program with the intent of improv- ing the overall quality of child care in these facilities." Two underlying factors in the decision to adopt registration were the issue of cost and an attempt to 28 bring more "underground" family day care operations under the regulatory umbrella. In practice, registration has lowered the per-home cost of regulation by reducing the level of state screening and monitoring and by dramatically in- creasing the number of homes falling under the regulatory umbrella. The out- come of DHR's recent evaluation of registration implementation reassures DHR officials that registration is working better than licensing previously did: the number of regulated homes has increased significantly; providers tend to view registration as an appropriate method of regulation; costs per unit of registration are lower than the costs for licensing; sample evaluations have in- dicated a high degree of compliance with minimum standards; and examina- tion of child abuse and neglect complaints do not indicate any greater danger to children under registration. Maria Chavez* and her husband are Mexican-Americans living in San An- tonio: both are bilingual in Spanish and English. He completed eleven years of school and now works as a bus driver. Mrs. Chavez, now 49, went to school through the eighth grade. She is a licensed family day care provider. Their 14- year-old daughter lives at home, but is at school for a good many of the hours when the day care children are there. When they are home, both her husband and daughter help look after the day care children. Mrs. Chavez cares for four children. Albert, who is nearly four, arrives each morning at ten and stays until five. His seven-year-old brother comes every day after school and also stays until five, doing his homework each day at the Chavez home. Because the two are brothers, their mother pays only $5 a day for both children. A third child, a toddler, is in full-time care, but her schedule is very irregular because her mother works in a hospital. Her mother pays $5 each day for her care. Mrs. Chavez's nine-month-old niece, Alicia, arrives early Monday morning and stays through to Friday afternoon— she stays with her aunt for the week and returns home each weekend. For this service, Maria Chavez also charges $5 a day. In all, she earns about $75 each week for her many hours of child care. Yet she regards family day care as a good source of income. Mrs. Chavez became registered with the Department of Human Resources almost two years ago. When she first heard of family day care registration by word-of-mouth, she called the Welfare Department herself because, she says, "I didn't want to break the law." A representative of the Fire Department visited her and checked for safety hazards. Shortly afterwards she heard from DHR— a representative telephoned her to say that she had "passed" and was now registered. The entire process took two months. Now Mrs. Chavez feels more comfortable: "I am more confident and more secure to care for the babies parents bring to me." Most respondents in San Antonio, including registered family day care pro- viders, reported satisfaction with the minimum standards for registered family day care. However, many doubt that DHR monitoring serves as an effective *In cases where profiles of individual providers are given, the names of caregivers and children have been changed. 29 control over quality. This is precisely the risk such a system takes because of the restriction of resources allocated on a per-home basis. However, the issue for Texas, and for all states, is the extent to which the State should be responsi- ble for protecting children in child care facilities. Studies conducted by DHR cited the following reactions. • People do not understand what is required of them in either the stand- ards or the registration packet. • The definition of registered family day care homes and the specifica- tion of the staff/child ratio standard is confusing to providers; most do not understand how many children they may care for. • The standards are too vaguely defined to be enforceable, especially regarding conditions for penalty or relocation of registration status. While Texas' registration system places a great deal of responsibility on the consumer, DHR's legal counsel does not believe it reduces the responsibility of the State to protect children in registered facilities. Although DHR does not currently provide services to the consumer and provides very limited services to caregivers, the legal counsel has recommended that DHR improve standards and their enforcement by educating providers on standards, by mounting an effective public education campaign to inform parents of registration and their role in monitoring standards, and by assuming more responsibility in training family day care providers. However, it is far from certain that these recom- mendations will be legislated and funded. Philadelphia Regulatory Context Pennsylvania statutes provide for the traditional licensure of family day care homes through initial home inspections and annual monitoring visits thereafter. In the Commonwealth of Pennsylvania, the office generally responsible for child care is the Department of Public Welfare, Bureau of Child Development. Licensing is administered through DPW regional offices. For homes under the umbrella of a sponsoring agency, the agency is licensed and delegated authority to * 'approve" homes, following the licensing regula- tions. All subsidized family day care is provided through agency-affiliated homes. For independent homes (those not attached to any agency), licensure is carried out by regional staff. Over the last several years, the Bureau of Child Development has focused some of its energy specifically on family day care regulations. New regula- tions, issued on April 4, 1978, cover center day care and family day care, as well as day care services for children with disabilities. One advocate observed that the new regulations, despite a level of detail which some have criticized as inappropriate for independent homes, are at least "clear, and it is possible to apply them equitably." Under the new regulations, family day care homes are limited to caring for six children, including the provider's own under the age of six, with a further limit of four infants/toddlers (0-36 months). The 90-item standards section of the regulations covers caregiver responsibilities and 30 qualifications; caregiver/child ratio; building and physical site; equipment; program for children; child health; staff health; food and nutrition; and transportation. Like the state of Texas, Pennsylvania uses its nearly $45 million in Title XX funds for child care to purchase subsidized care in center-based settings and more than half (57%) of these funds are allocated to the state's southeastern region which encompasses the city of Philadelphia. In Philadelphia,. visible day care is overwhelmingly center-based, and center-based care is overwhelm- ingly publicly subsidized (82%). These factors are reflected in the allocation of regional licensing office staff: 12 staff members are needed to monitor sub- sidized care; 2 workers handle all the region's family day care and provide center licensing. Regulatory Practices Overall, family day care — licensing, public awareness, and support ser- vices — is overshadowed by the heavily supported public school centers in Philadelphia and the need to achieve accountability for these center expen- ditures. Consequently, the regional office does not have the resources to carry out the full range of responsibilities associated with licensing. Family day care licensing itself is not adequately publicized, nor are unlicensed homes actively pursued. Independent homes are, in theory, visited initially on application and thereafter annually. However, officials at both the state and regional levels ex- pressed frustration at not having the resources to be able to conduct the annual visits in a timely fashion. To become a licensed provider in Pennsylvania, the potential caregiver must call or write the Department of Social Services and subsequently receive an in- formation package including a letter from the supervisor, a summary of regulations and a form requesting a visit. The caregiver is asked to attend a group meeting with other applicants to review regulations and to discuss prob- lems and questions about caregiving. Visits are then made to the home and chances are one in five that a change in the home will have to be made. Medical forms are required for each family member and one additional visit may be made before a certificate of compliance is mailed in which the caregiver is ' 'authorized to conduct and maintain a facility to provide a family day care home." Two months before the anniversary of the license, the caregiver will receive new application and medical forms and will probably be visited once before renewal is granted. Overall, it appears that family day care is a relatively little used form of care in Philadelphia and this is partial explanation for both the lack of resources available to independent caregivers and to that group's lack of organization and advocacy for its concerns. There is a clear line of division between those providers who operate under the sponsorship of one of Philadelphia's sponsor agencies and those who are independent providers. While the regulations are the same for all providers, sponsored caregivers are approved by their systems rather than licensed by the state. Sponsored providers are all offered training, while only 37 percent of the licensed independent providers reported having 31 received training. Overall, the independent provider in Philadelphia tends to be more isolated than her sponsored counterpart who has weekly contact with the agency. Joan Evans, a White woman, is 34 years old. Her two sons are in junior high now and don't need her care as they did when they were small. "I was bored," she says. "I wanted to work at home and be home when my children came back from school." Now she takes care of seven children whose parents work — six full-time and one after school. She works a ten-hour day and doesn't have time to be bored. Mrs. Evans takes care of a 9-month-old infant, a one-year-old, a 17-month-old toddler, two 3-year-olds and a 4-year-old. The 4-year-old's sister, who is in second grade, comes to Mrs. Evans' home every day after school. Their mother picks them up at 5:15. Mrs. Evans gets a lot of help from her family. Of her sons, she says, "They always wanted a little brother or sister and now there's a lot of kids around." Between her husband and her children, she estimates that she gets 15-20 hours of help with the day care children each week. They help out with just about every aspect of taking care of small children. When she heard about licensing from a friend who works for the State, Mrs. Evans applied because she didn't want to take care of other people's children illegally. Though she likes family day care in general, she does add, ''Sometimes I get tired of just seeing 3-year-olds. I'd like to talk to another adult just to keep my sanity." In Philadelphia and in Pennsylvania, two themes were consistently echoed in respondent views of federal involvement in family day care: leadership and flexibility. A common view expressed was that the federal government should first establish basic quality standards— not regulations— and then aid states in implementing these standards with sufficient flexibility to accommodate local needs in a politically appropriate manner. At the same time, there was an ap- parent disenchantment with the banner of accountability. "We spend all our time being accountable," yet "we owe more to the community than we can give them." That federal regulation regarding family day care should be directed at states and not at individual programs was a sentiment expressed by all respondents. Generally, it was felt that using federal monies for services rather than for regulation and accountability would prove more effective. One respondent best summed the central concern: "Are we worrying so much about accountability, eligibility guidelines and the like, that there are kids out there who can't get near our 'quality' services, because we won't let them?" Los Angeles Regulatory Context Funds for subsidized child care in California are administered by the State Department of Education through interagency agreements with the State Health and Welfare Agency which is the single state agency designated to receive all federal funds for social services. Title XX funds for child care have 32 been augmented substantially by state general funds through specific legislative efforts since 1972, designed to address a variety of child care needs and to expand the supply of child care in community-based settings. An important difference between California and the other two sites is the considerable investment of state funds in child care as a "buy out" of federal dollars to avoid the costs of federal regulation in subsidized care. Simply stated, Title XX monies are used for other social services, while the state has chosen to supplant federal monies with more state-funded child care. Day care subsidized by the state must meet state regulations which are more lenient and therefore less costly than those governing the purchase of federally subsidized care through Title XX. Under this "buy out" procedure, Chapter 344, Statutes of 1976 authorized $10 million to purchase alternative child care to test cost-reducing program features and to expand opportunities for parental choice. This authorizing legislation enabled the funding of approximately 30 family day care systems throughout California in 1976, thereby increasing the number of existing systems statewide by more than 100 percent. Seven of these are in Los Angeles. In 1976-77, the State Department of Education administered $93 million in day care funds of which $53 million were federally funded and the remaining $40 million were state funded. Overall, California has invested its subsidized child care dollars chiefly in center-based care as have Texas and Pennsylvania. In 1977-78, 93 percent of children receiving subsidies were cared for in centers, while the remaining 7 percent received care in family day care homes. However, AB 3059 initiated a trend in California to diversify the types of sub- sidized care available and to promote the growth of family day care systems and other community-based child care resources. In general, the regulation of family day care in Los Angeles is more stringently enforced than in either San Antonio, with its registration system, or in Philadelphia. The regulation of child care programs in California is the responsibility of both the federal and state governments. Programs funded through Title XX and Title IV-C operate under federal regulations; all other child care programs are under the regulatory authority of the state. The state delegates licensing authority to counties, many of which have returned the responsibility to the state because of the cost. Los Angeles County is one of the few large counties in the state to have retained the licensing function. Regulatory Practices Los Angeles County licenses three types of family day care homes: • homes providing care for up to five children, two of whom may be under the age of 2 years (including the applicant's children 16 years of age or under); • homes providing care for up to six children, between the ages of 2-16 years (including the applicant's children 16 years of age or under); and • homes providing care for up to ten children, between the ages of 2-16 years (including the applicant's children 16 years of age or under). The 33 applicant must have one helper/aide. The home is subject to fire and health codes applied to center facilities. Requirements concerning the physical characteristics of the home include space and equipment suitable for the ages of children in care, sanitation, cook- ing and toileting facilities, individual cots or beds, and adequate ventilation and temperature. Mrs. Gonzales resides in a predominantly Mexican-American neighborhood in Los Angeles County. At age 52, she lives with her husband in a two- bedroom apartment. Her three children, two boys and a girl, are grown but still live in the area. Mrs. Gonzales used to work part-time to supplement her household income, and in her extra time she spent as much time as she could with her daughter's first child. It was this relationship that started her off in the child care business — her daughter got a full-time job, and suggested that if she took care of other children, she could spend a lot more time with her granddaughter and still earn an income. The idea appealed to her. Mrs. Gonzales told us that she took care of three children. Of the three, one is her granddaughter, who is an infant of 9 months; the second is her niece, who is 14 months old; and another is an unrelated child who is 2-and-one-half months old. Although she has been taking care of children for five months, Mrs. Gonzales has never heard of the child care licensing requirements nor the applicable regulations. After a reluctant start in child care, Mrs. Gonzales thoroughly enjoys the time spent with the children and is very dedicated to her tasks. When asked if she did anything special for the children in her care, she responded, "I treat them as I would my own: I cradle them in my arms and often rock them to sleep; and I give them all my attention when they are here." The licensing process in California is relatively lengthy. Many counties hold monthly or semi-monthly meetings to familiarize potential applicants with licensing requirements and procedures. Each applicant is then required to sub- mit an application form, a pledge of nondiscrimination, a description of the physical features of the home, a report of physical examination, a report of tuberculosis test, and fingerprint cards for the applicant and her spouse. The applicant is then visited by a licensing worker who examines the home and, in counties where there is no training program, discusses the nutrition, health, and developmental needs of children, financial planning and methods of deal- ing with parents. When all forms have been received, including the results of a fingerprint check by the Bureau of Criminal Identification and Investigation (Department of Justice) in Sacramento, the licensing worker decides whether or not to issue a license. In some circumstances where there are already children in a family day care home and there is no obvious reason to deny a license, the county will permit the home to continue prior to the issuance of a license. In practice, a license is nearly always issued, because operators who are unwilling or unable to comply with the licensing requirements usually withdraw their application before this point. Among the reasons for denial of a family day care license, however, are prior conviction of the applicant or spouse for a felony involving 34 intentional bodily harm or a sex offense, falsification or withholding of infor- mation, inadequate facilities, or a determination that the applicant is physi- cally or emotionally unsuited to be a family day care provider. Notes 1. Some Aspects of Day Care Licensing at the State Level, Winifred A. More, Child Welfare League of America, Inc., 1957, p. 13. 2. Summary Statement on the Stage of Development of State Day Care Licensing, Edna Hughes, DHEW Children's Bureau Social and Rehabilitation Service, June 7, 1968, p. 4. 3. Day Care Licensing Study, Social and Administrative Services and Systems Association in conjunction with Consulting Service Corporation, August 1971, p. 4. An ACYF Com- parative Licensing Study, conducted by Lawrence Johnson and Associates, Washington, D.C., is currently in progress. That study will update state standards. 4. Children's Foundation, Family Day Care Food Advisory Project, Licensing Study, Washington, D.C., March 31, 1978, pp. 6-7. 5. Children's Foundation, op. cit. , pp. 37-38. 6. Texas Revised Civil Statutes annotated, Article 695c, Section 8 (a), subsection 1 (c). 35 Chapter 4 CHARACTERISTICS OF FAMILY DAY CARE PROVIDERS* The complex social and economic forces acting within each community establish an equilibrium between demand and supply in family day care. As a result of the complexity of the forces involved, the shape of this equilibrium differs from community to community. In some, young mothers choose to stay at home to care for their own and other preschool children, becoming family day care providers. In other communities, social and economic pressures force this same group of women to find employment outside of their own homes. In these communities older relatives, friends and neighbors generally provide the needed day care. The study design described in Chapter Two took such community variations into account. In order to describe this variety of family day care arrangements, caregivers in White, Black and Hispanic communities were selected by design from sponsored, regulated and unregulated homes. Subsequent analyses of the characteristics of the caregivers interviewed for this study have shown that many of the most significant similarities and dif- ferences among caregivers reflect differences in caregiver ethnicity or dif- ferences in the regulatory status of the family day care home. Because of the significant nature of these variations, we have taken the unusual step of presenting many of our findings in terms of the 3x3 table defined by caregiver ethnicity and regulatory status of the home. That is, many of the results of the study are presented in the format of Table 3 above. For many of our findings, presentation of a single ''average" picture would have distorted the meaning of the results. For example, we will show below that there are three clearly distinct groups of caregivers: young White mothers in their late twenties and thirties with their own children at home; women in their forties and fifties with at least one relative's child (often a grandchild) in care; and women in their thirties to fifties who care for the children of friends, neighbors and others in the community, but are not caring for the children of relatives. By presenting our results in terms of a basic 3x3 table, these com- plex study findings are clarified rather than masked. The caregiver profiles presented in the first section of this chapter cover a range of topics, including the reasons for becoming a caregiver and basic ♦Information in this chapter is condensed from the NDCHS Final Report, Volume II, Research Report, prepared by Abt Associates. 37 caregiver demographic data. Caregiver qualifications and issues relating to the stability of the day care arrangement are dealt with in subsequent sections. Profile of Family Day Care Providers The central figure in family day care is the provider. Her background, her motivation for providing care, her child-rearing attitudes and practices and other personal qualities shape the experiences of the children for whom she cares. Family day care providers are a socioeconomically and culturally diverse group. These women are young and old, married and single, rich and poor; yet they have one thing in common — they care for other people's children in their own homes. In this section, we present a profile of these providers, addressing the question, "Who is taking care of the children?" Women expressed a variety of motives for becoming child care providers. Some said that they had always loved children, had always worked with them and that caring for them in their own homes was the most natural of profes- sions (26.6%). Another 22.4 percent said they began taking in children because of the extra money they could earn. Still others responded that they had nothing else to do, or that they wanted to work but preferred to remain at home. Despite the varied reasons for starting to provide care, most family day care providers like their job and feel that it is permanent rather than temporary. Across sites, three-quarters of the providers interviewed did not intend to change jobs or stop working. Not surprisingly, sponsored and regulated pro- viders, who have gone through a certification process, most often think of their job as permanent; also, older, more experienced caregivers are more likely to think of their job in these terms. Almost 50 percent of the unregulated caregivers, on the other hand, are providing care on a short-term basis and foresee a time or circumstance when they will stop. Caregivers interviewed in the National Day Care Home Study ranged in age from 16 to 76; the vast majority, however, fell between 25 and 55 (see Figure 2). Across sites, the median age was 41 .6, indicating that although a large pro- portion of the caregiver population is composed of young women, often with young children at home, an equally large fraction of the population consists of middle-aged and older women who have already raised their families. Table 9 presents the median caregiver age by ethnicity and regulatory status. The most striking finding in this table is that White unregulated caregivers — White women providing informal care outside of the regulatory system — are substantially younger than any other group of providers. They are only slightly over 30 years of age, whereas every other, group of caregivers averages at least eight years older. Upon closer examination, we find that these White caregivers are by and large mothers who have chosen to become family day care providers while their own young children are at home. Subsequent tables will show that they tend to come from households with relatively high household incomes, and 38 Figure 2 Distribution of Caregiver Age (N = 784) 5 - 20 25 30 40 45 50 55 60 65 70 75 Years of Age Table 9. — Median Caregiver Age.* Sponsored Regulated Unregulated White 40.3 38.4 30.4 34.7 Black 44.5 49.4 46.9 46.6 Hispanic 47.5 39.9 44.8 44.8 44.5 42.3 38.4 41.6 Because the distribution of caregiver age is slightly skewed, medians, rather than means, have been used for presentation. that the income from caregiving represents for them only a small portion of total household earnings. Black and Hispanic caregivers, on the other hand, tend to be substantially older than the White caregivers, although the difference is most striking for unregulated providers, where there is approximately a 15-year difference— a real generational gap. Many of these Black and Hispanic women were caring for at least one relative along with a group of unrelated children. However, whereas among Whites this was often the caregiver's own child, among Blacks and Hispanics it was frequently a grandchild. Household incomes are lower 39 Histogram of Table 9 Median Caregiver Age A&* SPON REG UNREG for these women than for their White counterparts, and consequently family day care income is often a more significant contribution to the household. Two profiles of individual providers will illustrate the generation gap between White and non-White unregulated caregivers. Kathi Henderson is 34 years old and can hardly remember a time when she was not taking care of other people's children, although she has never been licensed as a family day care provider. She herself was married and pregnant with her first child at 18, when she took her first day care children. She ex- plains: "I wanted to be home with my kids, and it's important for the other kids to have love while their parents are at work .... My mother worked when I was young, and I always felt like an outsider." Mrs. Henderson, a White woman, has four children — two daughters, 16 and 13, and two sons, eight and five. The youngest, Peter, is still at home during the day. Here, he is the oldest: his mother cares for an eight-month-old infant, a year-old infant, an 18-month-old toddler, a two-year-old girl and a four- year-old boy. "Peter," she says, "gets along better with the little ones." He needs lots of assurance from her that he is special, though, with all the others around. Kathi Henderson contributes $150 a week to the household income. Her husband, a custodian, makes about $1 1,000. She's had a job only once during their marriage — for six months — and she prefers family day care even though she can't make much money at it. She generally charges $30 a week for each child, but will take less if a child's parents cannot afford that much. For this fee, she provides a lot of very personal service. Four-year-old Mike, for exam- ple, has asthma. He needs to get his medicine regularly, has special pillows, and it's important to keep him calm and happy. The youngest child is allergic to milk and drinks a soybean formula. The extra effort does not bother Mrs. Henderson; it is all part of the job, and she feels like she gets a lot of help from her two daughters when they come home from school. 40 Mrs. Johnson is a Black woman who heads an extended family consisting of her husband, her niece and her niece's two-and-one-half-year-old daughter. She became a family day care provider — an unlicensed one — after a neighbor suggested it to her. "I like the children around me all the time," she says. "They're all I have." Mrs. Johnson's husband, who has a sixth-grade educa- tion, doesn't have a job, so her income from child care is especially important to the family. In addition to her grandniece, who is always at home, Mrs. Johnson takes care of three day care children. All three children are Black. Two of them, 13- month-old David and 19-month-old Wanda, are in the home from 7:30 to 5:30, Monday through Friday. So during the day, Mrs. Johnson has three children under three in her home. In addition, she takes care of a five-year-old girl for three evenings a week. Not infrequently, one of the younger children also stays on into the evening. Mrs. Johnson, 61 years old, says, "I'm always on the go." Her husband and niece help out quite a bit though, as does a 14- year-old neighbor. In all, Mrs. Johnson makes about $65 a week as a family day care provider. This works out to a little over 56 cents per child per hour, although she varies her fees according to how much parents can afford to pay. Slight though this income is, it makes up about half of the"household's income— the remainder comes from Social Security and rent paid by Mrs. Johnson's niece. Mrs. Johnson considers her work as a family day care mother to be tem- porary, but she adds, "Each year I say I'm not going to babysit and then parents just ask me." Many women begin family day care by caring for the child of a relative. Fully 56 percent of all caregivers provide care during the day to at least one related child — their own child, a grandchild, niece or nephew (see Table 10). Relative care is especially common among unregulated caregivers. Over three- quarters of unregulated caregivers were found to care for a related child com- pared to only 35 percent of sponsored caregivers.* Table 10. — Percentage of Homes with Any Related Child * Sponsored Regulated Unregulated White 34 45 32 49 79 68 79 57 Black Hispanic 38 31 47 63 35 42 76 56 a Resident and nonresident; caregiver's own and other related children. *In the next chapter, on group composition, we point out that although many providers care for at least one related child, by far the preponderance of children in care are unrelated to the caregiver, except in unregulated family day care homes. That is, a typical arrangement might in- clude one relative and four unrelated children. 41 However, the pattern of relative care is one of the few with substantial unex- plained variation from site to site. Whereas in Los Angeles and San Antonio most unregulated caregivers we interviewed cared for at least one related child, in Philadelphia the pattern was quite different, with substantially less relative care. We were unable to determine the reason for this variation. The following profile illustrates a typical relative care arrangement. Lucy McCoy is a 54-year-old Black woman who works as a family day care provider. She began taking care of other people's children about five years ago, with no license, because her cousin needed day care and various other relatives also needed her services. These days, she is a licensed provider and takes care of four children of friends and neighbors. Her own granddaughter, Linda, who lives with her, is also at home during the day. Linda, age two-and- a-half, shares her grandmother with three toddlers 22 to 23 months old, as well as a six-month old infant. All of the children are Black. Three of the four day care children are from single-parent families. Mrs. McCoy reports that Linda would like to help her take care of the other children, but she's really too young. She does like having the other children around to play with, and they all get along very well. In fact, even though the day care children place demands on Mrs. McCoy's time, her own family appreciate her being a day care mother. Mr. McCoy, like his wife a high school graduate, works as a gardener at a country club. Together, they earn almost $9,000 a year, on which they support themselves, their two high-school children and their granddaughter. Mrs. McCoy set her own fee after discussing it with other caregivers she knew; she charges $25 a week for each child. Three years ago, Mrs. McCoy decided to obtain a license, after hearing about it at a day care meeting she had been invited to attend. She felt if she were going to be in the day care business, it would be better to have a license. It is important to her to get "reliable working parents, who want the best care for their children." The large majority of family day care providers in this study were married; across sites, three-quarters of those interviewed were currently living with a spouse (see Table 11). Although over 80 percent of the White and Hispanic caregivers were married, only 50 percent of Black providers were. This dif- ference is not surprising, however, given the large proportion of single-parent families within the Black population nationally. Table 1 1 . — Percentage of Married Caregivers. Sponsored Regulated Unregulated White 88 82 87 84 Black 55 53 49 52 Hispanic 86 92 72 80 73 75 71 73 Among providers who were married, the large majority ($1%) had husbands who were employed. Thus, a sizeable proportion of caregivers come from 42 Histogram of Table 1 1 Percentage of Married Caregivers SPON REG UNREG households in which there are two wage-earners, and caregiving is not the only source of income. Family day care providers spanned a broad range in household income. Many caregivers in our sample had household incomes under $6,000; indeed, this is the modal income category. Nevertheless, a small but important fraction had incomes in excess of $21,000 per year. Median household income across all study providers was found to be just over $10,000 per year. It is noteworthy that White caregivers in all sites were substantially better off economically than either their Black or Hispanic counterparts (see Table 12). Table 12. — Median Caregiver Income.* Sponsored Regulated Unregulated White $17,250 $12,750 $12,750 $13,500 Black $ 6,750 $ 9,000 $ 4,750 $ 6,750 Hispanic $ 9,000 $10,000 $ 6,750 $ 9,000 "Because the distribution of caregiver income is skewed, medians have been used for presentation rather than means. On average, only a small proportion of total household income of family day care providers comes from caregiving. Half the caregivers in the NDCHS sample relied on this resource for less than one-quarter of their total income. The notable exception to this pattern is found among sponsored Black pro- viders, many of whom are the sole support of their families. Almost half of these providers earned all of their family income by providing child care. Although total household income for many of these family day care pro- viders was extremely low, and for many, the provision of child care was the 43 sole source of income, only a small percentage of caregivers interviewed reported receiving any kind of welfare assistance. Among those providers whose total annual income was less than $6,000, only 15 percent relied on the government for any support. Thus, although caregiver wages are low (see Chapter Eight), family day care providers appear to constitute a self- supporting segment of the population. Caregiver Qualifications Three caregiver characteristics are often used as indicators of caregiver pro- fessionalism — education, experience and training. In this section the general distribution of these factors in the population is presented. In Chapter Seven, the relationship of these factors to caregiver behavior, the true indicator of caregiver professionalism, is explored. Of the three dimensions of caregiver qualifications described here, educa- tional attainment has traditionally been considered the least relevant index of competence for family day care providers. Nevertheless, it is still held by some to be an index of professionalism and thus is included in this profile of caregiver qualifications (see Table 13). Table 13. — Median Years of Education. Sponsored Regulated Unregulated White Black Hispanic 12.6 11.6 9.4 12.0 11.5 10.2 12.2 11.3 9.3 12.2 11.4 9.6 11.3 11.5 11.0 11.3 Although very few of the caregivers under study had a college degree (see Figure 3), the majority had completed high school (57%), and some had gone on to take at least some college courses. On the other hand, a full 19 percent of the caregivers had only an eighth-grade education or less. In keeping with the results of many other studies, the education of caregivers was found to vary significantly across ethnic groups. Hispanic providers had approximately two years less education, on average, than Black providers, and almost three years less than White providers. This reflects the fact that only 5 percent of White caregivers and 10 percent of Black caregivers had less than a ninth-grade education, but over 50 percent of the Hispanic caregivers never went past grade school. Moreover, in San Antonio, we interviewed several Hispanic pro- viders whose families had forbidden them to attend Anglo schools; conse- quently, these providers had virtually no formal education. Thus, although White providers had about one year more education on average than their Black counterparts, the major educational distinction among ethnic groups is that between Hispanic and non-Hispanic caregivers. Length of family day care experience is a commonly cited indicator of caregiver professionalism. Other things being equal, parents are inclined to 44 Figure 3 Distribution of Years of Education (N = 793) 40- 30- 20- 10 ^8 12 13 Years of Education choose the more experienced caregiver.* However, experience is somewhat dif- ficult to define. What should be included? The caregiver's experience in other day care settings? Her experience in raising her own family? Or only her ex- perience in family day care? Upon closer inspection, it turns out that very few family day care providers have had any day care experience other than in fam- ily day care. For example, only 7.9 percent of study providers queried had worked in a day care center, and an even smaller proportion had experience with other preschool or elementary school programs. Experience outside of family day care does not therefore appear to be a useful indicator of caregiver qualifications. Measurement of the variation in experience gained in raising one's own family had the opposite problem: not surprisingly, the vast majority ♦See CSPD's Parent Study Component Data Analysis Report (Volume IV of the NDCHS Final Report) for a full discussion of parents' perceptions of professionalism. 45 of caregivers had children of their own. Because this measure does not vary across providers, it, too, is not a very useful measure of experience. ' All of the caregivers we interviewed had some experience in providing family day care, if only for a short period of time, but length of this experience varied substantially across providers. Among the caregivers interviewed, some had less than one month's experience and a large proportion (25%) had less than one year of experience. Fifty percent had one to seven years of experience, and the remaining 25 percent had been providing care for 7 to 36 years. We will return to this issue of experience in Chapter Seven. As we shall show in Chapter Seven, the most important indicator of caregiver professionalism is training. Almost three-quarters of all sponsored caregivers have received some child care training, but this is true of less than one-third of all regulated providers and an even smaller proportion of unregulated providers. Because of the enormous variety of the training that caregivers received, it proved difficult to develop a single measure of this training. Some caregivers attended weekly classes given by their system or local schools, whereas others attended a single seminar. Further, we had no way of objectively assessing, after the fact, the quality of the training received. However, we were able to capture information about caregivers' perceptions of the usefulness of their training. Their sentiments were mixed. Some pro- viders felt that their experience in raising their own family taught them most of the skills necessary to provide child care and that training had done little to supplement this knowledge. On the other hand, many felt that the training program helped them substantially, especially by suggesting ideas for things to do and how to interact with children. Caregivers often emphasized that train- ing had taught them specific skills, good nutrition, bookkeeping, basic first aid, educational games and discipline techniques. Yet another group indicated that the supportive experience of meeting with other women and discussing the provision of family day care with professionals was the most helpful aspect of training, not necessarily the specific information taught by the instructor. Given the varied nature of the training received and the range of responses to this training, we were surprised to find how useful training turned out to be. The findings of the analysis on the relationship of training to caregiver behavior is presented in Chapter Seven. Summary of Caregiver Characteristics The previous sections paint a picture of three groups of caregivers: young White mothers in their late twenties and thirties with their own young children at home; women in their forties and fifties with at least one relative's child (often a grandchild) in care; and women in their thirties to fifties who care for the children of friends, neighbors and others in the community, but are not caring for their own or a relative's child. Whereas the first two groups tend to provide unregulated care, the last group of providers constitutes a large pro- portion of the regulated and sponsored caregivers. 46 The differences between younger and older providers extend, however, well beyond questions of their relationship to the children in care. The older caregiver tends to be more experienced but less educated than her younger counterpart and, if she is married, her husband also will have less education than that of the younger caregiver. Her husband is also most likely to be unemployed. Consequently, this older caregiver tends to have a lower total household income and earns a larger proportion of this income from child care. In contrast, there is a younger, better educated caregiver who has her own young children at home and does not provide care for her relatives. She is likely to be married and to have a husband who is employed and better educated than both herself and the husbands of older caregivers. Her total household income is generally higher than that of older providers and is not re- quired to support the household. Stability and Continuity in Family Day Care An issue of vital concern to consumers of family day care is whether the caregiver will provide a lasting source of care. Parents feel that it is important to provide their child with a secure, constant day care environment; for the parents themselves, such a secure relationship ensures that work and daily life will not be disrupted. The term stability will be used to refer to the provider's tendency to continue to provide care over time; the term continuity will be used to refer to the child's tendency to remain in the same family day care home over time. Conventional wisdom holds that family day care homes are quickly established and short-lived, and that the day care children themselves are tran- sients in short-term, stop-gap arrangements. To address this issue, we collected data in one site (Los Angeles) at two time points one year apart. By comparing data at the two time points, we were able to assess stability and continuity of care over the course of the year. Stability of Providers One-quarter of the caregivers recontacted for a second time one year later were no longer providing child care.* Although caregiver attrition was ap- parently not related to ethnicity, the regulatory status of the caregiver seems to have made a major difference: almost half of the unregulated caregivers had stopped taking care of children within a year of the initial interview,** com- pared with 13 percent of sponsored caregivers and 16 percent of regulated caregivers. Thus, a formal commitment to family day care, made explicit when •There were 68 caregivers who could not be recontacted in Los Angeles. Whereas over 80 per- cent of sponsored and regulated caregivers could be found, only two-thirds of the unregulated caregivers could be contacted again. This represents further evidence of the relative instability of unregulated care. **Note, however, that most of these unregulated caregivers had been taking care of children for a year or more before the initial interview. 47 caregivers obtain the sanction of regulation, is associated with an increased tendency to continue taking care of children. When the group of caregivers who were still caring for children was com- pared with the group no longer providing care, a number of differences emerged. These differences, taken together, seem to show a contrast between a career caregiver and a caregiver who is providing care on a short-term basis. Caregivers who were no longer caring for children were significantly younger and less experienced than those who were still caring for children. The former had a mean age of 39 years and the latter 45.5 years. Further, caregivers who had stopped providing child care had a mean of 3.5 years of day care ex- perience, compared to 6.8 years for caregivers who were still caring for children. Caregivers who were no longer caring for children also had significantly smaller day care groups, with an average group size of 2.8 children, compared with 4.2 children for those still providing care. By con- trast, average hourly fee and average age of day care children did not differ between groups. Nor were group differences found in the number of the caregiver's own young children in the home, the percentage of day care children related to the caregiver, her years of schooling, household income or day care training. Moreover, caregivers themselves are a good source of information on this issue. In the original data collection, caregivers were asked whether they con- sidered family day care to be permanent or temporary work. The great major- ity of these caregivers (77%) accurately predicted whether they would continue in family day care.* More than 60 percent described their work as permanent and were still providing day care one year later. An additional 15 percent described their work as temporary, and one year later had stopped taking care of children. Only one caregiver in five (23%) did not accurately predict her later behavior. Sixty percent of these had described their work as temporary; yet one year later at the time of the telephone survey, they were still providing child care. (In some cases, these caregivers may have interpreted "temporary" to mean an indefinite period longer than one year.) The remaining 40 percent of these caregivers responded that their work was permanent yet had stopped taking care of children by the time of the follow-up. In sum, only 9 percent of the caregivers originally questioned had unexpectedly ceased to provide day care. Typical perceptions of family day care have held not only that it is short- lived, but also that its informality is accompanied by constant shifting of group size and group composition. NDCHS data, however, showed that the basic characteristics of home operations were stable over the year. Only one characteristic, mean hourly fee, was found to have changed for the 175 •Note that these caregivers as a group may be very stable in their involvement in child care. These caregivers participated in four NDCHS interviews and that participation may have been associated with greater overall stability. 48 caregivers who continued to provide care. In a year of high inflation, fees in- creased significantly from $.72 an hour to $.80. Changes in regulatory status were also examined among caregivers still pro- viding family day care in order to discover whether unregulated caregivers tend to become licensed or remain unregulated. Over the course of the year, only an insignificant number of licensed caregivers gave up their licenses, and an equally insignificant number of unregulated caregivers became licensed. Thus, regulated and unregulated caregivers were found to be very distinct popula- tions with little movement between them. Continuity of Care for Children Children in the NDCHS entered family day care at ages ranging from birth to 10 or more years. Length of time in care is often expected to vary with the age of the child when the arrangement began. For example, it is sometimes hypothesized that the older a child is when he enters care, the shorter his stay is likely to be. Another hypothesis relates time in care to the developmental stages of a young child's life. Within this framework, children who enter day care as infants may be expected to graduate to a group care setting when they become preschoolers, thus consistently limiting their time in family day care. To test these hypotheses, analyses were performed to identify age cohorts of children with distinct patterns of length of time in care. For example, these analyses assessed whether all preschool children had similar expectations for length of time in care, and whether as a group, their length of time in care dif- fered from that of infants and toddlers. Analyses have shown that there are three child cohorts with distinct patterns of attendance: children starting care between birth and six months of age, children starting between 6 and 12 months of age and children starting at one year of age or older. The youngest group of children, those starting at less than six months of age, remained in care the longest. In this age group, 90 percent of the children were still in care after one year and 75 percent after two years. The point at which half of the youngest group had left care was four years. In comparison, among children who started at age 6 to 12 months, half were still in care 2.25 years later. The group of children who started care at one year of age or older had the highest attrition rate; half of them were gone in less than two years. Provision of care by a relative was significantly related to length of time in care and to the age of the child at start of care. Children being cared for by an aunt, grandmother or other relative (excluding mothers) entered care younger than nonrelative children. In the case of children cared for by relatives, 54 per- cent were less than one year of age on entering care; only 32 percent of the children entering a nonrelative home were this young. Furthermore, in the sample as a whole, children in the care of a relative stayed in care significantly longer than children not related to the caregiver. Half were still in care after more than five years, compared to two years and two months for unrelated children. 49 Chapter 5 DAY CARE CHILDREN: GROUP COMPOSITION* Each day millions of children are cared for in family day care homes, more than in any other out-of-home setting. This chapter asks the question, "Who are these children?" The typical family day care home has one caregiver in charge of a group of children. The composition of this group — number of children, relation to the caregiver, age mix and so forth — has long been assumed to have a profound effect on the character of the home, including the kinds of experiences the children have, the overall atmosphere of the home and the burden on the caregiver. The importance of group composition is reflected in widespread interest among parents, providers, program planners and policymakers in issues related to group size and age mix of children in family day care homes. Parents' interest is expressed in a preference for the home-like atmosphere and grouping in family day care, particularly for the very young and for school- aged children needing a few hours of supervision in the afternoon. Developmental specialists, educators and other child advocates voice concern about the number of children for whom a single adult can care while still pro- viding adequate individual attention to children who need it. For policymakers, group composition is viewed as a potentially powerful proxy for quality which can be regulated and monitored to protect children from harm and enhance their development in the family day care environment. No single number adequately describes group composition — the distribution of the children in a family day care home. On the one hand, one may examine this distribution by analyzing the number of children enrolled in each home. This construct — enrollment — is the most frequently regulated home character- istic and is believed to influence a wide variety of home processes including caregiver burden and the quality of care delivered. An alternative description of this population can be based upon the ages of children in care. Enrollment is only a crude characterization of caregiver burden; clearly, the ages of the individual children enrolled can be shown to exert as much, if not more, influence upon home processes than their sheer numbers. We will therefore present the age composition of the NDCHS sam- ple of children — their ages, the age mixes found in family day care homes and the relationship between age mix and enrollment. •Information in this chapter is condensed from the NDCHS Final Report, Volume II, Research Report, prepared by Abt Associates. 51 Group Size The variety of ways in which group size can legitimately be measured reflects the complexity of an issue which at first glance appears quite straightforward. The group size of a particular home is a function of the method by which the number of children and the number of caregivers are computed. Analysis of the concept of group size is dependent upon two important measurement issues. • Which children and adults should be included in measurements of caregiver/child ratio and group size? • Should these measures be based on observations or are lists of in- dividuals present during the day sufficient? These issues will be addressed in stages. First, different measures of group size will be described. The single simplest measure — the number of nonresident children enrolled in the home — will be used as an index of size, and then an alternative measure that includes resident children (usually the caregiver's own children) will be discussed. Regulations for family day care have typically placed a ceiling on the number of children that may be placed in a family day care home. The 1968 Federal Interagency Day Care Requirements (FIDCR), the federal purchasing requirements for subsidized care which were in place during the course of this study, restricted family day care homes as follows. • For children from infancy through 6 years: no more than 2 children under 2 and no more than 5 in total, including the family day care mother's own children under 14 years old. • For children from 3 through 14 years: no more than 6 children in- cluding the family day care mother's own children under 14 years old. Thus the FIDCR enrollment limit was either five or six depending on the ages of children in care.* The definition incorporated in the FIDCR is but one of many that appear in day care regulations and in the literature. These definitions vary in the way the number of children in care is calculated. Children of different ages tend to be counted differentially, with younger children counting more heavily; the caregiver's own children are often treated differently from nonresident children in the home. In Los Angeles, for example, at the time data for this study were collected, two different standards were being used. One included in total enrollment the caregiver's own children under 14 (as in the FIDCR); the other counted only her children under 7 years. Often definitions ignore another, equally serious issue — the time at which the enrollment count is taken. For example, a count of all children who are ever present in the home may not reflect the actual number in the home at any given time. Moreover, it is important to consider whether, under certain cir- cumstances, it is preferable to measure the caregiver/child ratio, which takes * Revised federal regulations approved March 16, 1980 allow for additional school-aged children 'hen no infants are present. 52 into account the predominantly casual, but sometimes regular, presence of secondary caregivers. The number of nonresident children enrolled in the home is most often used as the indicator of the size of the family day care group. Across all 793 NDCHS homes, the number of nonresident children per home ranged from one to 22 with a median of 3 (see Figure 4). Most homes (90%) had six or fewer children enrolled, dispelling the frequently held misconception that a substan- tial proportion of family day care homes have inordinately large groups. This finding closely parallels findings from earlier studies of family day care, most notably Keyserling's national study and studies carried out in Pennsylvania and Oregon. 1 The number of nonresident children per home was found to vary signifi- cantly across caregiver ethnicity; Hispanic homes had one child less, on average, than non-Hispanic homes (2.7 vs. 3.8). (See Table 14.) Enrollment also varied significantly by the regulatory status of the home; sponsored homes (4.3 children) were larger on average than regulated homes (4.0), which were in turn larger than unregulated homes (2.8). However, the relationship of regulatory status and group size was not consistent across all sites. (See Table 15.) In Los Angeles, sponsored homes were twice as large as unregulated homes; enrollment in regulated homes fell closer to that of unregulated homes. Unregulated homes in San Antonio are also extremely small, but enrollment in regulated homes is, on average, larger than that in sponsored homes by one child. In stark contrast, Philadelphia's unregulated homes are larger than either their sponsored or regulated counterparts. Table 14.— Mean Number of Nonresident Children Per Home By Caregiver, Ethnicity and Regulatory Status. Sponsored Regulated Unregulated White Black Hispanic 4.3 4.6 3.8 4.4 3.9 3.4 3.1 3.3 2.1 3.8 3.8 2.7 4.3 4.0 2.8 3.5 The particularly large enrollments found in Los Angeles sponsored homes seem to reflect the lack of exclusive use agreements between most systems and their providers.* Thus many of the children in sponsored homes were recruited by the caregiver herself and not by the sponsoring agency. These non-exclusive use providers are free from sponsoring agency enrollment limits, and many care for almost as many nonsponsored children as sponsored children.** On the other hand, all sponsored care in San Antonio (there was only one agency *Under exclusive use agreements, a caregiver may only enroll those children who have been referred by the sponsoring agency. **In addition, licensing regulations were suspended for several California systems operating under state funding. This allowed them to enroll more children than currently permitted by state licensing regulations. 53 Figure 4 Distribution of Number of Children per Home (793 homes) 26- 24- 22- 20- 12- 10- Median = 3.0 Mean = 3.5 Standard Deviation = = 2.5 r^^i 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 in existence at the time of the study) and most sponsored care in Philadelphia was based on exclusive use arrangements between agency and provider. The majority of system providers in these cities are therefore restricted to enrolling only as many children as the agency is able or willing to place. Enrollment in regulated homes follows a different pattern. Both licensing and registration increase caregivers' access to referral sources, making it possi- ble for providers to care for larger numbers of children. In fact, many caregivers become regulated because "you can't advertise or get any [child care] referrals without a license." In Los Angeles and San Antonio, an in- creased access to families in need of care through these more formal mechanisms may account for much of the enrollment difference observed be- tween regulated and unregulated providers. Among regulated providers, the highest enrollments are found in San Antonio. 54 Table 1 5. —Mean Number of Nonresident Children Per Home by Site, Caregiver, Ethnicity and Regulatory Status. Sponsored Regulated Unregulated LOS ANGELES White... Black . . . Hispanic White... Black . . . Hispanic White... Black . . . Hispanic 5.7 4.2 3.0 3.9 6.9 3.2 2.8 4.0 3.9 2.9 2.2 2.9 5.4 3.7 2.7 3.7 SAN ANTONIO — 4.7 2.9 3.8 3.6 4.7 2.3 2.9 3.5 3.7 2.1 2.6 3.5 4.4 2.4 3.2 PHILADELPHIA 3.3 4.1 3.8 3.7 3.6 4.2 4.9 4.2 — — — — 3.5 4.2 4.4 4.0 Unregulated homes tended to be smaller than either their sponsored or regulated counterparts; the exception to this rule, however, was found in Philadelphia, where unregulated Black homes were among the largest in that site. One explanation of this result is that enforcement of family day care regulations is rather weak in Philadelphia; as a result, unregulated providers may feel that they can easily take in many children without suffering adverse consequences. This pattern may also be explained by other factors peculiar to Philadelphia. Our interviews with members of the day care community in Philadelphia sug- gested that in contrast to the other two sites, family day care there may be a relatively unused form of child care. It is possible that in this and other north- ern industrial cities, one impact of dense and often stressful urban living is a reduction in the number of people willing to care for children in their homes. This would, in turn, increase the demands on those who do provide care to meet the needs of additional families; consequently, enrollments may expand in homes that are not subject to regulation.* Although the above profile provides a grounding in the distribution of basic enrollment, as well as in the variation of this measure across the design factors, it does not convey the complexity of this issue when the caregiver's own children are considered. •There also are indications that Philadelphia's Black providers may have a greater professional commitment to family day care than do Whites. A comparison of Whites' and Blacks' household income shows that Philadelphia's Black providers rely on the income earned from family day care much more than do Whites. It may be that the younger and wealthier White providers supply care for fewer children and on a more temporary basis than do Blacks in that city. 55 Resident Children Related to the Caregiver Children related to the caregiver fall into two groups: those who reside in the caregiver's home; and those who come to the home for daily care. Naturally, most related children who live in the caregiver's home are the caregiver's own, but they may also be grandchildren, nieces, nephews and cousins. The discus- sion of enrollment in the previous section incorporated nonresident-related children but excluded resident children. This section focuses on the caregiver's own children and others who live in her home. There were relatively few resident children under age 14 living in the homes of the caregivers interviewed for the NDCHS (see Figure 5). Almost half of the 793 providers had no children under 14, 25 percent had one child, and 17 per- cent had two children under 14. Overall, slightly more caregivers had school- aged children (aged 7-14) than children under seven (average of 0.7 and 0.6, respectively). An interesting pattern emerges upon examining the distribution of resident children under seven (see Table 16). Unregulated White providers consistently have more young children of their own at home than any other group (average of 1.1 across sites).* This issue was raised earlier when it was pointed out that these providers are also, by a large margin, the youngest group of caregivers. They appear to care for family day care children only when they have young children of their own at home. As there are more job opportunities for these women, it is often not practical for them to remain as caregivers when their own children are no longer at home during the day. On the other hand, their relatively higher incomes often allow these women the choice of staying at home when their children are young. Table 16.— Mean Number of Resident Children Under Seven Per Home. Sponsored Regulated Unregulated White Black Hispanic 0.5 0.4 0.3 0.6 0.2 0.5 1.1 0.5 0.6 0.8 0.4 0.5 0.4 0.5 0.8 0.6 One White caregiver stated, "I was thinking of going back to work but the problem arose with my children. My youngest boy is only two and I hated the thought of someone else taking [care of] him. So I had a choice: either go out and work again or stay home and take care of children with my own. I decided to stay home." An attempt to define group size by including resident children meets with a critical difficulty. Which of the caregiver's own children should be included in group size? Those under five? Those under seven? Those under seven only if •Note, too, that Black providers as a group tend to have the fewest young children; as they also constitute the oldest caregiver group, this finding is not surprising. 56 Figure 5 Distribution of Resident Children Under 14 (793 homes) 30 20 10- Median = 1.0 Mean = 1.3 Standard Deviation = 1.3 Number of Resident Children Under 14. they are present during the day? Those under 14? No commonly accepted delineator exists for determining who should be included in such a measure. For completeness, then, and in an effort to determine who should most ap- propriately be counted, various measures of group size were constructed for this study. Obviously, overall group size increases as more of the caregiver's own children are included in the measure. Except for unregulated care, where total enrollments are quite small, none of the differences turns out to be very large, and if one restricts the enrollment measure to include only those of the caregiver's children under seven who are home during the day, the difference 57 between enrollment as defined in the previous section and enrollment plus resi- dent related children is minimal: the number of children increases on average by less than one, from 3.5 to 3.8. Relationship Between the Number of Related Children and Enrollment In the previous chapter we pointed out that 57 percent of all caregivers care for at least one related child. In the previous section we looked at resident children. In this section we will look more closely at the group of nonresident children by dividing them into those who are related to the caregiver and those who are not. This distinction can be used to investigate further the structure of enrollment within a home and can provide some insight into a caregiver's motivation for operating a family day care home. The average numbers of nonresident related children in homes of various types are displayed in Table 17. The overall mean, 0.5, suggests that there is not a high density of nonresident related children in day care homes especially in sponsored and regulated care. This may appear surprising to some, con- sidering the number of homes in which care for related children is provided. However, except in unregulated homes, the related child is frequently only one of several children in care. Table 17. — Mean Number of Nonresident Related Children Per Home. Sponsored Regulated Unregulated White Black Hispanic 0.2 0.2 0.3 0.2 0.3 0.3 0.4 0.9 1.2 0.3 0.5 0.8 0.2 0.3 0.8 0.5 Nonresident related children, however, are significantly more likely to be enrolled in unregulated homes than in either sponsored or regulated homes, and Hispanic and Black providers are more likely than White providers to care for these children. This predominance of nonresident related children in unregulated homes with Hispanic or Black caregivers can be compared to the predominance of resident related children in unregulated homes with White caregivers. First, children (either resident or nonresident) who are related to the caregiver appear more frequently in unregulated homes. This suggests that many unregulated caregivers manage day care homes because they wish to pro- vide care for relatives. Regulated and sponsored caregivers, on the other hand, appear to have this motivation less often — they are more likely to be "in the business" of caregiving. Second, unregulated caregivers of different ethnicities are caring for different sets of related children. White caregivers have their own children in care; Hispanic and Black caregivers have young relatives in their charge. The extent of the care of relatives is demonstrated further in Table 18, where the percentages of related nonresident children are displayed by site. Over half 58 the children in Black and Hispanic unregulated homes in Los Angeles and San Antonio are related to their caregivers. In fact, the vast majority of these ar- rangements consist of grandmothers caring for their nonresident grand- children. This provision of care seems to reflect both the informal and familial environment that is characteristic of much unregulated care and, more specifically, the important role still played by the extended family in these communities. In both Los Angeles and San Antonio, Mexican-American families continue to provide for themselves many supports not fully available in the larger community. Many of these caregivers who are grandmothers are unregulated because they do not perceive themselves as family day care pro- viders. Rather they state, "I am only taking care of my grandchildren." Table IS.— Percentage of Nonresident Related Children. Sponsored Regulated Unregulated LOS ANGELES White... Black . . . Hispanic White Black Hispanic White... Black . . . Hispanic 6.0 7.3 9.6 2.7 10.9 11.1 8.2 40.0 57.0 4.8 14.3 25.8 7.6 5.9 28.2 12.0 SAN ANTONIO 0.0 7.1 6.6 8.1 10.4 19.0 55.4 53.7 11.1 32.5 32.0 4.1 7.8 41.2 22.5 PHILADELPHIA 2.7 0.8 10.6 4.3 8.3 6.2 7.4 3.9 .4 6.4 7.0 5.2 One somewhat atypical result revealed in this table is the lack of relative care among Philadelphia's unregulated Black caregivers. No satisfactory explana- tion is available for why these providers are not caring for their relatives' children in the same proportions as Black unregulated caregivers in the other sites. Ages of Children in Care Although family day care homes in the NDCHS served children of all ages, toddlers (19-36 months) constituted the largest group (32%). Other ages were almost equally represented. This finding offers a sharp contrast with center care, where preschool children are the largest single age group. Of the 2812 children enrolled in NDCHS homes, 23 percent were infants, 32 percent were toddlers, 22 percent were preschoolers and the remaining 23 percent were 59 Histogram of Table 1 9 Age Distribution of Children in Care By Regulatory Status 100 SCH AGE PRESCH TODDLER INFANT SPON REG UNREG schoolers (see Table 19). This result corresponds directly to that obtained from interviews with the parents of children in care. These parents reported that family day care was the preferred type of child care for their toddlers, in-home care was preferred for infants, and center care and other group settings were preferred for preschoolers. Thus, toddlers are the modal age group in family day care homes, just as preschoolers predominate in centers. Table 19.— Age Distribution of Children in Care by Regulatory Status. Sponsored Regulated Unregulated Infants (0-18 mos.) Toddlers (19-36 mos.) . . . . Preschoolers (37-60 mos.). Schoolers (61 + mos.) . . . . 18.5% 25.4% 22.6% 22.9% 42.0% 32.5% 24.4% 31.7% 23.7% 23.2% 20.8% 22.5% 15.7% 19.0% 32.2% 22.9% 100.0% 100.0% 100.0% 100.0% The age distribution of children in family day care has undergone a shift over the past decade. Studies conducted during the early 1970s found larger numbers of preschoolers and smaller numbers of infants and toddlers in family day care. 2 The changing trends probably reflect both the increasing popularity of center day care for preschoolers and the growing numbers of mothers with very young children who have recently been entering the work force. As Table 19 shows, the population of children served varies considerably by the regulatory status of the home. Toddlers are clearly the modal age group in sponsored homes; across sites, these children account for over 40 percent of 60 the population in sponsored homes. Balancing this increased proportion of toddlers is a decrease in the number of infants and schoolers served in spon- sored care. Regulated homes resemble sponsored homes in that toddlers are also the modal age group; the proportion of toddlers, however, is substantially smaller because more infants and schoolers are found in these arrangements. In direct contrast, schoolers are the modal age group in unregulated care, and infants also constitute a sizeable fraction of this population. The substantial proportion of toddlers and the coincident smaller fractions of infants and schoolers in sponsored care can be explained by two factors relating to the economics of placement practices in family day care systems. First, although many systems would like to enroll more infants, regulations which limit group size more stringently if there are several infants produce a strong economic disincentive against enrolling infants. In homes where infants are placed, the agency must limit the number of other children placed, thus decreasing the overall number of children who can be served and consequently the caregiver's income.* Second, school-aged children are infrequent in spon- sored homes because of the part-time nature of their care. Since caregiver in- come is based upon the total number of child-hours, it is not profitable from the agency's perspective to accept schoolers because these children are gen- erally in care for fewer hours than younger children but still occupy one of the day care slots as specified in the day care regulations.** Thus, placing schoolers once again limits the number of additional children who can be served. These disincentives, however, are not as strong in regulated homes, even though they are restricted by the same set of enrollment ceilings, simply because these regulations are not as strictly enforced as they are in sponsored homes. As a result, there are sizeable numbers of infants and schoolers in regulated homes, although toddlers remain the largest group. Finally, unregulated homes are not influenced by this set of disincentives. Hence, the need for day care for infants and school-aged children, which is not being met in sponsored and regulated homes or in centers, is being met by the unregulated family day care market. The discussion thus far has focused on the average age composition of the major regulatory categories of family day care homes. In concluding, let us review age composition from the perspective of the individual caregiver, on a home-by-home basis. Almost all caregivers restricted the age mix of children served to minimize their own burden and to increase the number of age- appropriate playmates for their own children and for the children in their care. Nearly one-third (31%) of providers cared for children in a single age group (Table 20). Homes caring only for infants or only for toddlers were common (10% each), although homes caring for only preschoolers or only school-aged children were less frequent. More than half of the remaining caregivers ♦Only rarely are reimbursement rates for infants higher than those for other children. **The recently proposed new federal day care requirements, recognizing this disincentive for caring for schoolers, increases the number of school-age children permitted if there are no infants in the home. 61 (39.2%) cared for children from two age groups; infant/toddler and tod- dler/preschooler combinations were the most common. Indeed, the in- fant/toddler combination is the modal grouping of children found in family day care homes. Statistical examination of the age mix of the study homes con- firmed that homogeneous groups (for example, infants only or infant/toddler) tend to be more common than heterogeneous groups (for example, infant/school-aged) . Table 20.— Distribution of Age Group Configurations. [All Sites; N = 793] Configuration Number of Homes Percentage of Homes Total ONE AGE GROUP I.... T.... P.... S.... 79 80 37 49 10.0 10.1 4.7 6.2 245 31.0 TWO AGE GROUPS I-T .. 104 13.1 T-P.. 70 8.8 P-S.. 46 5.8 T-S.. 32 4.0 I-P .. 39 4.9 I-S .. 21 2.6 Total 312 39.2 THREE AGE GROUPS I-T-P 87 11.0 T-P-S 56 7.1 I-P-S 25 3.2 I-T-S 30 3.8 Total S Total 198 25.1 FOUR AGE GROUPS I-T-P- 38 4.8 793 100.1 I = Infant P = Preschooler T = Toddler S = School-aged child. Compliance with Regulations At the outset of this study there was considerable concern about the level of family day care homes' compliance with federal and state regulations. This concern was expressed most frequently about group size, in light of its pre- sumed relation to quality of care. Because of the minimal level of enforcement nationally and the large proportion of unregulated homes, it was commonly held that a large proportion of homes were out of compliance. Study findings show that this is not the case. The great majority of homes were in compliance with both state and federal group size requirements. Since individual states 62 have their own ratio requirements, caregiver compliance will be examined with respect to these local regulations. In Los Angeles, regulations are similar to the federal requirements, except that the local requirements include all the caregiver's own children under 16 (the FIDCR include caregiver's children under 14). At the time of the study, 47 percent of sponsored homes, 77 percent of licensed homes and 78 percent of the unlicensed homes were in compliance with California's ratio requirements. The reason for the apparently low level of compliance among sponsored homes is that many of those homes were exempt from ratio requirements under State Assembly Bill 3059 which provided a state-financed alternative to federal day care programs. Thus, while these homes appear to be out of com- pliance, they are, in fact, complying under a separate, less stringent set of standards. In Philadelphia, 95 percent of sponsored homes, 73 percent of licensed homes and 74 percent of unlicensed homes were in compliance. Here, as in Los Angeles, licensed and unlicensed homes are very similar in their level of com- pliance. The high level of compliance of sponsored homes is due to the ex- clusive use arrangements used by sponsors. These arrangements place strict limits on the number of children cared for in family day care homes, limits that are enforced by the system staff. In San Antonio, the registration system allows more children in a home than in either of the other sites. One would expect, therefore, that compliance would be higher here than in the other sites. Indeed, all of the sponsored homes were within registration limits. Only 18 percent of registered homes and 4 percent of unregistered homes had more children than the registration limits allowed. Overall, the regulatory system appears to be equally effective in all sites. Sponsored homes generally complied strictly with state regulations, due to the affiliation with and supervision by the sponsoring agency (except in Los Angeles, as discussed). Regulated and unregulated homes across sites, however, were very similar in their level of compliance. The implications of this similarity in size of child care groups, from a regulatory perspective, are quite clear: with respect to group size, most caregivers are self -monitoring. In general, caregivers — knowingly or unknowingly — conform to the standards applied within each state. Notes 1. Keyserling, M.D. Windows on Day Care (New York: National Council of Jewish Women, 1972). Peters, D.C. Day Care Homes: A Pennsylvania Profile (University Park, Pa.: Pennsylvania State University, 1972). Unco, Inc. A Profile of Federally Supported Day Care in Oregon, prepared for the U.S. Department of Health, Education and Welfare, 1974a. 2. Unco, Inc. A Profile of Federally Supported Day Care in Washington, prepared for the U.S. Department of Health, Education and Welfare, 1974b. 63 Chapter 6 PARENTS OF THE CHILDREN IN CARE* Chapter Four began our description of the major day care actors with an in- troduction to day care providers; Chapter Five continued with a description of the children in care. This chapter concludes with the last of the principal fam- ily day care actors, the parents. The focus of the interviews with parents was on their day care needs and preferences and their assessment of their current day care arrangements. Parents represent the demand side of the day care relationship, just as pro- viders represent the supply side. No study of family day care could be complete without assessing the reasons for which the parent selected it and the reasons behind the parent's continuing or terminating the relationship. Policymakers, advocates and researchers alike must consider their positions in light of parents' concerns in addition to the needs of the child in care, which so often receive primary consideration. Parents' Background Characteristics The parents of children in the National Day Care Home Study sample were selected because their children were enrolled in the family day care homes that were originally chosen for the study. These parents do not represent a random sample of the population but instead are drawn from those parents who have access to family day care homes in the selected study neighborhoods. Often their families lived in the same neighborhoods as the caregivers. To understand the needs and preferences of these parents, it is first important to gain some general idea of who the users of family day care are. In general, the income of the parents in the NDCHS sample (median $12,000-$ 15,000) was lower than the national average (median $16,000), largely because the White parents in the sample had lower incomes than their counterparts nationwide. This is consonant with the fact that caregiver homes were selected primarily from lower-income areas within the study sites. Within the NDCHS sample, however, Whites had a higher income level than Blacks or Hispanics: over twice as many Whites as Blacks or Hispanics, for example, ♦Information in this chapter is condensed from the NDCHS Final Report, Volume IV, Parent Study Component Data Analysis Report, prepared by the Center for Systems and Program Development. 65 had incomes over $21,000. At the other end of the scale, only 6 percent of White parents had incomes under $6,000 — as opposed to 18 percent of Hispanics and fully 25 percent of Blacks. The NDCHS sample was also slanted toward occupations of lower status. Moving upward on the occupational ladder, a smaller proportion of the sam- ple appears at each step: 37 percent unskilled, 29 percent skilled trades and sales, 25 percent lower level managerial/professional, and only 7 percent up- per level managerial/professional. The only clear relationship between paren- tal income and home setting is an expected one. Parents of children in spon- sored care had lower average incomes than parents of children in regulated or unregulated care. This is because sponsored care is most often subsidized. Over 60 percent of parents in the NDCHS sample were married (or infor- mally married). Twenty-five percent were divorced or separated, and 15 per- cent were single (including widows/ widowers). Of married parents, only about one-fifth used sponsored care, the least popular type of care for these parents because, in general, they are not eligible for day care subsidies. Among single parents, on the other hand, nearly half used sponsored care. Again, this con- centration of single parents among the users of sponsored care is not surpris- ing: the incomes of single-parent families tend to be less than those of two- parent families, and sponsored care is frequently subsidized. Finally, there was a slight tendency for the users of nonsponsored care to have longer tenure in their neighborhoods; this greater familiarity with their neighborhoods may be what allowed them to tap sources of day care that are not part of a formal sponsoring network. Compared to national averages, the NDCHS sample was composed of more small families (one or two children) and fewer large families (three or more children). The concentration of families with only one child was especially high. Well over half of the children in the sample were under three years old. This finding is consistent with the findings of earlier studies that parents tend to choose family day care for children under three and center care for preschoolers aged three to five. When parents' reasons for seeking day care were probed, the great majority (86%) indicated that child care enabled them to work. There was, however, a difference across ethnic groups: fully 94 percent of Whites but only 88 percent of Hispanics and 77 percent of Blacks gave this reason. Parents' reasons for selecting family day care were also probed; more than half cited financial con- siderations, special attention for the child or unavailability of center care. Although most respondents worked typical daytime hours, a sizeable minority worked at night or had rotating or variable work schedules, suggesting a need for flexible child care arrangements. In general, there was a strong tendency to prefer care in the child's own home for children under one year, to select fam- ily day care for one- to three-year-olds, and a more structured environ- ment — center care, nursery school or kindergarten — for older preschoolers. Eleven percent of parents gave their child's age as the main reason for selecting family day care. 66 Overall, about half of the parents interviewed had placed their child in the type of day care that they preferred. The level of congruity between preferred and actual mode of child care was greater in sponsored settings than in regulated or unregulated care. In addition, a far higher percentage of Black respondents (61%) had other day care preferences for their children than did Whites (36%) or Hispanics (48%). In general, Black parents voiced a strong preference for center care, with its more apparent educational component. The majority of parents in the NDCHS sample had located their current caregiver through some personal source — a relative, friend or neighbor — and most of these parents stated that they would follow the same route again if seeking child care in the future. In general, advertising and agencies played only a small role, although California's Resource and Referral Centers (which maintain lists of child care providers) were a significant source of information for parents in Los Angeles. About one-fifth of NDCHS parents felt some reservations when placing their child in family day care and stated that they had made compromises in deciding to engage a particular family day care pro- vider. The most common sources of concern for parents were the caregiver's pets (which frightened some children) and issues of personality — finding a caregiver with whom they felt comfortable leaving their child. Parental Expectations and Degree of Satisfaction Parents were asked about the most important characteristics of the family day care home in which they would leave their child. Most frequently men- tioned (by 20% of parents) was the reliability of the caregiver and a close se- cond was that the child acquire linguistic and cognitive skills (26%). Other often-mentioned needs were that the child be well cared for and receive emo- tional support, that nutritional needs be met and that the environment be safe and clean. When questioned about the special needs of their child, very few parents mentioned any medical or emotional problems. About one-fifth of parents responded that special care for infants was needed. Most striking, nearly half of all parents (43%) felt that their child had a special need to be with other children — indeed, so many parents mentioned this factor that it ap- pears to be a very common concern in this era of the shrinking American fami- ly. On the whole, parents seemed to feel that their children's needs are met about 90 percent of the time. What features did parents look for in evaluating family day care homes for their own use? A convenient location seemed to be important to parents, but was rarely a problem. Parents were also generally satisfied with the physical characteristics of homes. Food was the most often mentioned problem area — 14 percent of parents felt that caregivers sometimes served inap- propriate foods (e.g., junk food). Moreover, about one-third of all parents supplied all or part of their child's food themselves— considering the low parental fees paid per child, this provision of food is an indirect way of sup- plementing the caregiver's income as well as insuring the quality of the food provided to the child. 67 Parent's reactions to the number and age mix of children in care are in- teresting in light of their concern with cognitive gains and with exposure to other children. Three times as many parents stated that there were too few children in their child's family day care group as complained that there were too many.* And three times as many parents felt that the children in the group were too young as felt that they were too old. Parents appear to be concerned not only that their child have the company of other children, but also that these others be children from whom their child can learn. When asked about important qualifications of family day care providers, most parents (82%) rated experience with children as more important than for- mal education, although Black and White parents seemed to value education more highly than Hispanic parents. This general lack of regard for formal education as evidence that a provider is "good with children" reflects parents' greater respect for concrete experience. It may also reflect a perception that the highly educated provider may be frustrated by the low earnings and non- professional status of family day care and may therefore not provide the most positive environment for children. About three-quarters of the parents interviewed were not comfortable with substituting another caregiver for their child's regular caregiver in case of caregiver illness or vacation. However, parents with children in sponsored homes were generally more flexible on this issue, suggesting that in this more businesslike environment, either caregivers are seen as more interchangeable or parents trust the sponsor to find a suitable substitute. Parents' attitudes about their child's daily activities in the family day care home were probed to discover the extent of parental satisfaction and also parents' preferences for different types of environments. Only about one- quarter of NDCHS parents were dissatisfied with their child's daily activities; about half of these parents explicitly sought greater emphasis on conceptual and linguistic skills. NDCHS data on parents' preferences are difficult to inter- pret: when asked to choose between two hypothetical homes, parents chose an unstructured environment over a structured one by a ratio of 3 to 2, but they also chose a "learning" environment over a "play" environment by a ratio of 3 to 1. The phrasing of these forced-choice items may well have influenced parents' responses; however, what does show clearly is parents' pervasive con- cern with cognitive development.* * Finally, parents were queried about the availability of special ser- vices — evening and weekend care, care for a sick child — that are often cited as a major advantage of family day care over center care. Surprisingly, responses suggest that the flexibility of a typical family day care home is not as great as has often been assumed. Only about one-sixth of parents reported that evening *This suggests that it would be beneficial to parents as well as to caregivers to improve day care referral sources so that somewhat larger homes can be maintained. **The importance to many parents of the child's development of conceptual and linguistic skills was reflected at several points in the parent interview. Chapter Seven discusses differences in the encouragement of these skills from one day care setting to another. 68 and weekend care were available — somewhat less than expected. Nearly 20 percent, however, reported that their caregiver provided care to a seriously ill child, a service rarely if ever provided in centers. Other special services, such as parent education or family counseling, were more readily available through sponsored homes than through regulated or unregulated homes, as expected. Although many parents claimed to need such institutional services when they were available, these services did not figure prominently when parents listed their requirements for family day care. The needs that parents mentioned spontaneously were more fundamental ones — a reliable caregiver and a good learning environment, for example. In a similar vein, parents were questioned about unexpected benefits of family day care or, on the other hand, expectations that had gone unmet. Many parents had placed their child in care in spite of an underlying convic- tion that no one could care for their child as well as they, and in particular that no one else was "good enough" to give their child the special attention that he deserved — a natural enough feeling for parents leaving their child in another's home. However, nearly half of the parents who felt there had been unexpected benefits from the family day care experience stated that their child received much more individual attention from the caregiver than originally anticipated. This is a strong endorsement of the home atmosphere of family day care and matches the findings of our own in-home observations of the behavior of caregivers with the children in their care. (Chapter Seven describes the results of these observations.) Smaller numbers of parents reported that the caregiver's personality had impacted beneficially on their child, that nutri- tional provisions were surprisingly good, that food had been supplied at no cost or that the child had learned more physical skills from the caregiver than they had expected. Only a few parents reported that certain of their expecta- tions of family day care had not been met. Of these parents, 20 percent in- dicated that they had hoped for a greater emphasis on cognitive development, yet another indication of the strength of parents' ambitions in this regard. Other unmet expectations were so scattered that the results are not inter- pretable. Parents' responses to another set of questions, while striking, are very hard to interpret. When asked whether their child had ever had a "bad experience" in any family day care setting (including previous arrangments), 11 percent of parents reported that this was so. The most commonly mentioned bad ex- perience was an injury to the child (slightly more than 2%); 2 percent cited in- adequate supervision; fewer than 2 percent said that their child had been left unattended; and the same proportion said that their child had been physically abused. It is impossible to determine from these data whether such "bad ex- periences" should be considered a normal part of the typical child's develop- ment, or similarly, whether the same sorts of incidents might have occurred in the child's own home. What then did parents perceive as benefits reaped by their children from the family day care experience? Over 60 percent of NDCHS parents cited their 69 child's social growth, suggesting that this care arrangement meets parents' demands for their child to be with other children. One-third of parents felt that their child's linguistic and conceptual skills had grown in family day care (although, as noted above, about 13 percent of parents would have preferred a greater emphasis on the acquisition of such skills). About one-fifth of parents mentioned the homelike atmosphere of family day care as a special benefit to their child. Another indirect indicator of parents' satisfaction with family day care is their perception of their child's attitude toward the provider. Three-quarters of parents in the National Day Care Home Study reported that their child had loving feelings for his or her caregiver; another 22 percent said their child's at- titude was friendly, though not loving. Only 3 percent of the children were considered by their parents to be indifferent to their caregivers, and no parents reported that their child disliked the provider. Finally, parents in Los Angeles were asked what aspects of family day care they felt should be regulated. Many parents said that such features as caregiver health, home safety, number of children and child health were suitable for regulation, but parents agreed that the internal social dynamics of the family day care home should not be regulated. Caregiver/Parent Communication Because family day care can be a uniquely personal child care arrangement, the nature of the caregiver/parent relationship represented an important ele- ment in the family day care picture. The majority of parents (63%) had not been acquainted with their child's caregiver before making arrangements for family day care. Not surprisingly, prior friendships between parent and pro- vider were much more common in unregulated and regulated care than in sponsored care. About half of the parents interviewed described their relation- ship with the provider as one of casual friendship, about one-third said they had a close personal friendship, and the remainder described their relationship as businesslike. Predictably, the closeness of the relationship was inversely related to formality of the day care structure. Close relationships most often occurred in unregulated arrangements; this reflects both the large amount of relative care that pervades unregulated arrangements and the friendship net- works that often underlie unregulated care. Close personal friendships in the family day care relationships were also more frequent among Hispanics than among Whites or Blacks. Some respondents felt that a close personal relationship was advantageous in dealing with issues between parent and provider — such as differences in at- titudes or behaviors with respect to childrearing. Other respondents, however, felt that such closeness made it very difficult to discuss problems or parental dissatisfactions, especially if the provider was a relative. The demands of fam- ily or friendship sometimes conflicted with parents' needs as consumers. In general, however, parents felt that they were in agreement with their caregiver on important aspects of childrearing. Although parents and sponsored 70 caregivers were not typically friends, these arrangements often enjoyed other resources that facilitated communications between parent and caregiver. In sum, the level of involvement between provider and parent is unusually high in family day care as compared with that in other day care settings. The high correspondence of attitudes about the relationship between parents and caregivers indicates that they are mutually interested in pursuing a relationship focused on the child but grounded in a friendship between the adults. Thus, the mutual concern and interest expressed by caregivers and parents indicates that family day care provides a unique opportunity for all concerned: the child, the parent and the caregiver. 71 Chapter 7 OBSERVATIONS IN FAMILY DAY CARE HOMES* In the preceding chapters we have described the family day care partici- pant — caregiver, parent and child. With this introduction, it is now possible to examine the core of the family day care arrangement — the day-to-day in- terplay between caregivers and the children in their care. In-home observations were conducted to answer such questions as the following. What kinds of ex- periences do children have in family day care and how do they go about their daily activities? What kinds of things do caregivers do with children and are these activities appropriate given the ages of the children? What is the in- terplay between caregiver and child — for example, what kinds of behaviors do caregivers facilitate and control? How do children behave both in their interac- tions with other children and in their solitary play? We were especially concerned with how the interaction patterns and the kinds of activities available for children in the family day care homes would be influenced by other characteristics of the home — the number of children pre- sent and their ages, caregiver characteristics such as age, education, experience and training, the regulatory status or administrative structure of the home and so forth. The National Day Care Home Study observations in family day care homes were conducted by trained observers using two in-home observation systems developed initially for this study by Dr. Jean Carew and SRI Interna- tional — the Carew/SRI Adult Behavior Codes and Child Codes. All observa- tions were carried out by SRI. The observation data were collected in two contexts within a home: in a natural situation as caregivers and children went about their normal daily ac- tivities and in experimentally structured situations where a common set of ac- tivities was presented in all homes. The former allowed comparison of homes based on observations of what the caregiver and children typically did during the day, and the latter allowed comparison of homes on the basis of a stan- dardized set of activities. Observations were used to characterize the caregiver ♦Information in this chapter is condensed from the NDCHS Final Report in Volume II, Research Report, prepared by Abt Associates and Volume III, Observation Component, prepared by SRI International. 73 and child independently and to provide detailed information on the interac- tions between caregivers and children. The observations also provide an objec- tive assessment of the family day care environment which can be used to sup- plement the caregiver's own report on her interactions with the children in her care. In the natural situation, caregivers and children were observed as they went about their usual activities for approximtely two hours during each of two mornings. In using both the Adult Behavior Codes and the Child Codes, behavior was sampled every 20 seconds for periods of five minutes at a time.* During each hour of observation, the observer completed eight five-minute segments. The aspects of caregiver behavior covered by the Adult Behavior Codes in- cluded: • how the caregiver interacts with children — adult strategies likely to facilitate or restrict child activity, such as teaching, helping, convers- ing, controlling or playing; • the type of activity that the caregiver facilitates, restricts or is otherwise involved in with the children; • the caregiver's expression of positive, negative or neutral feelings toward children; and • the caregiver's use of language in her interactions. The Child Codes were used to summarize child behaviors. These codes focus on the following four dimensions: • the child's activity; • the person with whom the child is interacting; • the child's participation in conversation; and • child affect. Examples of child activities that can be coded by the Child Codes instrument include language, fine motor activities, gross motor activities, conversation, work and watching television. In addition, such child behaviors as prosocial behavior, antisocial behavior and distress can be recorded. Observation Variables For the analysis of caregiver and child behavior, observation variables were constructed from the codes on the instruments. Basic concepts from child development theory were used as a framework for the development of the child observation variables. A congruent set of observation variables was developed to describe adult behaviors. The following list provides examples of adult behaviors and activities in- cluded in major observation variables. ♦Every 20 seconds, the observer noted on a specially prepared form what the caregiver was do- ing during a 3-second interval. This random sample of behavior allows us to generalize to the caregiver's behavior over the entire period. 74 • Adult involvement with children. Adult involvement ranges from ac- tive involvement (teaching, playing and participating in children's ac- tivities) to indirect involvement (supervising children) to no involve- ment (household chores and recreational activities). • Adult facilitating behaviors. With these behaviors, the adult promotes or participates with children in prosocial, affectionate and comfort- giving behaviors; language/information activities, fine motor struc- tured, fine motor exploratory and gross motor activities, music, dance and dramatic play; television; and physical needs activities. • Adult control activities. Adult control techniques range from positive routine control measures, for which explanations are given, to strict negative control measures. • Adult affect. Adult affect ranges from positive (laughing and smiling) to neutral (neither positive or negative) to negative (angry, hostile, sad). • Adult interactions with babies and older children. Adult interacts with a baby or older child. This variable is included to account for the caregiver's interactions with children outside the age range of this study (12 to 59 months). The following are examples of variables developed to describe child behavior. • Persons with whom the child is interacting. Child is engaging in an ac- tivity alone, with an adult, with a young child or group of children, with a baby or older child. • Content of child's activities. Child is engaging in prosocial, affec- tionate and comfort-giving behaviors; language/information activities; fine motor structured, fine motor exploratory and gross motor ac- tivities; music, dance and dramatic play; television; physical needs or antisocial activities. • Child's participation in a conversation. Child initiates, receives, or responds to conversation or to task-related comments; child uses Spanish or English. • Ch ild 's active engagemen t in an activity. • Child affect. Child affect ranges from positive (laughing and smiling) to neutral (neither positive nor negative) to negative (angry, hostile, sad). Quality of the Observation Data Since the observation data systems used in the National Day Care Home Study were developed specifically for this research, there were no previous studies to establish their reliability. Accordingly, a series of reliability studies was conducted. The first study was designed to answer two related questions. Would White, Black and Hispanic observers perceive the same event in the same way? Did it make a difference if the person observed was of the same or 75 different ethnicity? This study showed that ethnicity was not a factor in the ac- curacy and consistency with which observers coded behaviors of children and adults. A second group of substudies was undertaken to determine how stable the observations were over time: from hour to hour, from morning to afternoon, across two adjacent days, and over a period of four weeks. In each case very high reliability was established for both adult and child observations. For nearly every type of behavior, reliabilities of 90 percent or better were reported. In addition to the special substudies, observers in the NDCHS were required to maintain 80 percent accuracy with the coding system. Observers were tested against a criterion videotape at the end of training and again during field operations. Several of the substudies also involved assessment of inter- observer reliability in the field whereby two observers coded activities in the same home simultaneously and the completed observations were checked for consistency. As a result of these reliability studies, the observation system was shown to be psychometrically sound and the data of consistently high quality. Descriptive Analyses One of the primary objectives of the observation component was to describe how caregivers and children spend their time. For example, we wanted to know what proportion of caregivers' time is spent in direct interaction with children and how that time is distributed among different types of behavior. How much teaching, playing, and helping is there? How often do caregivers facilitate and control the children's behavior? In what ways? Because this was the first large-scale study of family day care to use this ap- proach, there were no norms to indicate what the "typical" distribution of the caregiver's time should be. Patterns of similarity and variation found in the behavior of caregivers among sponsored, regulated and unregulated homes help to develop an idea of what can be expected of caregivers in these settings. The same is true for children's behavior. The observation component was designed to describe the proportion of children's time spent in direct interac- tion with the caregiver, with other children and in independent activities. Within this broad context, we wanted to know how their time was distributed in developmentally important ways. How often did children engage in cognitive, social and physical activities? To what extent did they converse with adults and each other? What distinct behavioral patterns could be observed among children of different ages? Caregiver Behavior One of the most striking findings to emerge from the National Day Care Home Study is the high level of involvement with children exhibited by family day care providers. Overall, family day care providers spend nearly two-thirds of their day in child-related activities. On average, across all types of settings, ethnic groups and size of home, caregivers spend approximately half of their 76 time (46%) in direct interaction with children (Table 21; Figure 6). In addition, they spend another 17 percent of the day either supervising children or prepar- ing for them. Table 21 shows that 41.3 percent of the caregiver's time is spent interacting with the one- to five-year-old children in her care. This is in line with previous research on family day care, which also demonstrated that most caregivers spend a substantial proportion of their time in contact with the children. It also reflects parents' statements about the nature of the care that their children receive. Furthermore, this time seems to be spent in appropriate ways; that is, it mirrors the needs of children in care as indicated by their ages. One of the most frequent interactions with children in the one-to-five age range is teaching, which occupies 13.9 percent of the caregiver's time. In the remainder of the caregiver/child interactions, play/participation and helping are about equal in frequency, occupying 7.8 percent and 8.9 percent of caregivers' time, Table 21 .—Distribution of the Caregiver's Time* Percentage of Time Direct Involvement with Children Interaction with one-to five-year-olds: Teach 13.9 Play/participate 7.8 Help 8.9 Direct 3.7 Converse 3.3 Control 3.7 41.3 Interaction with babies (> 1 year) 3.8 b Interaction with school-aged children. 1.0 b Negative affect with any child c 0.3 Total Direct Involvement 46.4 Indirect Involvement with Children Supervise or prepare 16.5 Non-Involvement with Children Primary Caregiver: Converse with adults 6.3 7.8 19.4 Recreation alone Housekeeping Out of range 1.3 34.8 Secondary caregiver interaction 2.2 Total Non-Involvement 37.0 Grand Total 99 9 a This picture is based primarily on morning observations and does not include periods in which the children are napping or eating, or periods when more school-aged children are present. b These figures may be misleading as they stand. These percentages were calculated for all homes, even though caregivers were observed to interact with babies in only 35 percent of all homes and with school-aged children in only 16 percent of all homes. Con- sidering only homes in which babies are present, the caregiver spends about 1 1 percent of her time with babies. Considering only homes in which school-aged children are present, the caregiver spends about 6 percent of her time with them. Because of sampling requirements for observations, homes with babies and school-aged children are underrepresented in these data. c Negative affect occurred so infrequently that it could not be analyzed separately for each of the direct interaction variables but was instead aggregated into a single category of less than 0.3%. 77 Histogram of Table 21 Distribution of the Caregiver's Time: Interaction with One- to Five-Year-Olds TEACH PLAY/PAR HELP DIRECT CONVER CONTROL respectively. The amount of teaching and play in the homes in our study sug- gests that previous descriptions of family day care as lacking in stimulation for young children are not valid for our sample. Directing and controlling are substantially less frequent than other caregiver behaviors, each accounting for only 3.7 percent of the observed behaviors. Previous studies also found a minimal amount of restrictive behavior on the part of family day care pro- viders. Infants and school-aged children were only found in a subset of these homes. Whereas the observation instrument makes fine distinctions in the in- teractions of caregivers with one- to five-year-olds, their behavior with younger and older children is aggregated into coarser categories. Averaged across all homes, we found that the caregivers spent 3.8 percent of their time interacting with babies and one percent interacting with school-aged children. (See footnote b to Table 21 for a further explanation of these figures). Because negative affect was such an infrequent occurrence in family day care homes, it could not be analyzed separately for each of the interactions between caregivers and children noted above. Instead, all occurrences of negative affect were aggregated into a single measure. The reader may observe in Table 21 that, even in the aggregate form, the caregiver demonstrated 78 Figure 6 Distribution of Caregiver's Time negative affect with the children in her care in only 0.3 percent of the observa- tions. On average, the caregiver spent slightly over half of her time not actively in- volved with the children in her care. This time can be divided into two categories, indirect involvement and non-involvement. Indirect involvement consists of supervising the children (without interacting but with the potential of interacting with the children) and making preparations directly related to the care of the children. Together these activities comprise 16.5 percent of the caregiver's time, leaving her apparently uninvolved with the children for ap- proximately one-third of the time. When the caregiver was not occupied with the children, she spent her time either talking to other adults (6.3%), entertaining herself by reading or watch- ing television (7.8%), doing housekeeping (19.4%), or out of range of the observer (1.3%). Of course, even when she is not directly involved with the children, the caregiver can often respond to indications that something is amiss. In most cases, she was in close physical proximity to the children, being out of observable range only 1 .3 percent of the time. The final category in Table 21 relates to secondary caregivers. In some of the homes, especially larger regulated and sponsored homes, some portion of the 79 observations were taken up by helpers — secondary caregivers. On average, this constituted such a small proportion of the total observations (2.2%) that no useful analyses were feasible. Thus in subsequent analyses these secondary caregivers were dropped. Having described the distribution of the caregiver's time for the sample as a whole, we turn now to a consideration of similarities and differences in ob- served patterns of behavior for homes of different regulatory status. In fact, some of the most interesting study findings are implicit in comparisons of caregiver behavior across settings. Here we treat these comparisons in a purely descriptive manner, noting in which settings behaviors are relatively more fre- quent. The present discussion is consistent with the results of other analyses presented in subsequent sections and with the sense we have obtained of family day care through the caregiver interview process. Table 22 illustrates the striking differences between observed caregiver behaviors in unregulated, regulated and sponsored homes. Although all set- tings showed a substantial amount of caregiver/child interaction, the dif- ferences among settings are statistically significant. Unregulated homes showed the lowest level of caregiver interactions with one-to-five-year-olds in care. Regulated homes showed a somewhat higher level of interaction, and sponsored homes showed by far the highest level. The differences between unregulated and regulated homes were generally small; half of the difference in the level of interaction is in the frequency of helping behaviors, which ac- count for 8.2 percent of the caregiver/child interaction in unregulated homes and 9.5 percent in regulated homes. Sponsored homes, however, were substantially different in the amount of teaching that occurred. Whereas 12.1 percent of the caregiver's time in unregulated homes and 12.8 percent in regulated homes was spent in teaching, 17 percent of the caregiver's time in sponsored homes is spent in this man- ner — a very large difference. This is especially noteworthy in the light of find- ings from the NDCHS Parent Component and earlier research that many parents would prefer a day care environment which emphasizes teaching ac- tivities. Expanding upon these differences among settings, the bottom portion of Table 22 is devoted to the activities that caregivers facilitate with children while they are interacting with them. This table shows that sponsored caregivers facilitated much more language/information, structured fine motor activities and music/dance activities. Also, there was somewhat more gross motor activity in these homes and less watching of television. This implies a more preschool-like setting with structured activities for the children. On the whole, regulated and unregulated caregivers seem very different from spon- sored caregivers, but similar to one another. On the other hand, regulated and sponsored caregivers were similar with regard to the time spent in supervision and preparaton for children, whereas unregulated caregivers spend somewhat less time in these activities. Further- more, unregulated caregivers spent substantially more time uninvolved with 80 Table 22. — Distribution of the Caregiver's Time by Regulatory Status. Sponsored Regulated Unregulated Direct Involvement with Children Interaction with one- to five-year-olds: Teach 17.0 12.8 12.1 Play/participate 8.6 7.3 7.3 Help 9.3 9.5 8.2 Direct 3.9 3.8 3.8 Converse 3.4 3.3 3.0 Control 3.6 3.8 3.5 Subtotal 45.8 40.5 32.9 Interaction with baby 3.4 5.3 2.5 Interaction with school-aged Children 0.8 0.7 0.9 Negative emotional affect 0.2 0.2 0.4 Total Direct Involvement 50.2 46.7 41 .7 Indirect Involvement with Children Supervise/prepare 18.4 17.2 14.2 Noninvolvement with Children Converse 5.5 5.8 6.8 Recreation alone 4.8 5.7 13.0 Housekeeping 17.9 17.8 22.0 Out of range 1.7 1.4 1.2 Subtotal 29.9 30.7 43.0 Total 3 98.5 94.6 98.9 Activities Facilitated: Language/information 10.6 8.4 7.8 Structured fine motor 6.1 3.1 3.5 Work 1.4 1.3 1.3 Physical needs 8.2 8.9 8.1 Dramatic play 1.1 0.8 1.2 Music/dance 2.1 0.7 0.4 Television 1.8 2.6 2.0 Exploratory fine motor 1 .2 1 .0 0.9 Gross motor 2.2 1.8 1.5 Total facilitation 34.7 28.6 26.7 Positive Affect 6.0 5.6 4.6 a Recall that observations of helpers, averaging 2.2% overall, were deleted from the analysis. This accounts for total distribution of caregiver's time equalling less and 100%. the children. Forty-three percent of the unregulated caregiver's time is spent apart from the children, as opposed to about 30 percent for both regulated and sponsored caregivers. In unregulated homes, this means that, on average, 26 minutes of every hour were spent uninvolved with children whereas only 18 minutes of every hour were spent in this manner in sponsored homes. This set of findings suggests that the unregulated caregiver is somewhat less child-focused than the regulated caregiver and much less so than the sponsored caregiver. She spends more time than either attending both to her own needs and to her household's while the day care children are present. The sponsored caregiver's added involvement with the children is apparent in several ways: there is more teaching, more play/participation, more supervision and 81 Histogram of Table 22 Distribution of the Caregiver's Time by Regulatory Status 1-5 YR BABY SCH AGE PASSIVE NON INV preparation and less housekeeping and solitary recreation. Further, when we look at the additional activites that sponsored caregivers facilitate— language/information, structured fine motor, music/dance and gross motor— we see behaviors which are often stressed in the child care training that sponsored providers receive, but that few regulated or unregulated pro- viders do. This suggests that caregiver training may make a difference. This impression is supported by the study's other data sources, including caregiver interviews and observer summaries (recorded at the end of observation ses- sions). Finally, to put these findings into context, it is important to stress that NDCHS observers and interviewers were consistently impressed by the care that they saw regardless of regulatory status of the home. Conclusions, therefore, about the relative frequency of behavior from one setting to another are not intended to imply that unregulated care is poor care. However, many parents may well feel that sponsored care provides a more structured type of experience which they want for their children. Others may value a more in- formal care setting. Child Behavior Whereas, in general, there is only one caregiver in each home, there are usually several children in care. Observations of children were divided among the younger children (one- to three-year-olds) and the older children (three- to 82 Histogram of Table 22 Distribution of the Caregiver's Time by Regulatory Status: Interaction with One- to Five-Year Olds >r m- /*/ TEACH PLAY/PAR HELP DIRECT CONVER CONTROL five-year-olds). Where possible, one child in each age range was observed in each home; however, not all age groups were represented in every home. Therefore, the number of children observed in each category is considerably less than the number of caregivers observed: 303 homes and caregivers were represented, but only 253 younger and 161 older children were observed. Furthermore, because the children were grouped into such broad age spans and because they show rapid development during this age period, the variance in child behavior is often so large as to require multivariate statistical analyses to determine where there are actual differences in the behaviors between one care setting and another. In this section, however, we restrict ourselves to a simple descriptive analysis of the major characteristics of children's behavior in order to look at basic profiles. It is useful to compare the frequencies of behaviors for younger and older children, as this reinforces our sense that the observation system was responsive to the kinds of behavior one expects to see in children of these ages. In the socioemotional domain, for example, one notes more affectionate behavior among the younger children as well as more distress, both typical of this age group. Older children demonstrate more prosocial behavior and more frequently control the behavior of a younger child, again as one might expect. In the cognitive/linguistic domain, the older child exhibits more language/information activities, more dramatic play, more fine motor struc- 83 tured activity and more conversation with other young children. The younger child exhibits more fine motor exploratory behavior alone. The most common activities in both age groups involved fine motor skills such as puzzles, water play or manipulating small toys which were considerably more frequent than any other type of behavior. This picture is in contrast with some previous research indicating TV watching as the most frequent activity in family day care homes. Activities in the cognitive/linguistic domain were much more fre- quent than watching TV. In the physical/motor domain the younger child, as expected, has more physical needs to be met. The younger child also looks around, watches older children's activities or otherwise monitors the environment more than does the older child. Both younger and older children spend much of their time in independent activity; that is, not directly interacting with either the caregiver or another child. Younger children spend 62.6 percent of their time in this manner, whereas older children spend 51 .9 percent — less than younger children but still more than half of their time. The younger children spend 14.7 percent of their time directly interacting with the caregiver, and older children spend 11.8 per- cent of their time in this way. There are many fewer direct interactions among the children, however. Younger children interact only 1.7 percent of the time with older children, and older children 5.1 percent of the time. Although it is generally agreed that the caregiver's interactions with children are of para- mount importance to the children's family day care experience, the child's utilization of that large portion of time spent in independent activity is an equally important concern. Multivariate Analyses The descriptive profiles presented in the previous section set the stage for multivariate analyses of the observation data. The descriptive analysis asked, "What was the process like in family day care homes?" The multivariate analysis asks, "How might differences in caregiver and child behavior be ex- plained?" The descriptive profile alone suggested that regulatory status of the home is strongly associated with differences in behavior; the multivariate analysis went further by examining the observation data simultaneously across multiple factors of interest, assessing the relative importance of each in describing caregiver and child behavior. The simultaneous consideration of multiple factors afforded important ad- vantages over a variable-by-variable examination of the observation data. The variables under consideration were, in addition to the design variables, a set of factors such as caregivers' qualifications or group size which were not con- trolled in design but potentially related to caregiver and child behavior. Limiting the analysis to individual relationships between each variable and behavior would have been inappropriate — first, because the primary interest was in how the set of factors explained behavior; and second, because many of 84 the factors were confounded (i.e., not independent of each other). For in- stance, if caregiver behavior were associated with both caregiver training and the regulatory status of the home, the fact that training and the regulatory status are correlated means that these associations cannot be interpreted in- dividually; the effects could not be attributed solely to either factor. Any given behavior might be due to either or both of these elements. As many of the fac- tors of interest were confounded in the study design and/or correlated in the real world,* it was necessary to use techniques that could consider factors jointly and that could take into account the pattern of confounding. Multivariate analyses, although they could not completely disentangle related factors, did allow us to construct and test several different sets of factors which were selected to minimize confounding within a set. The results of the analyses are organized in terms of the principal factors of interest: group composition of the home (enrollment and age composition), regulatory status of the home (unregulated, regulated, sponsored), caregiver characteristics (background and qualifications), and child characteristics (own child in care, related children in care). The focus is on meaningful paterns of relationships across the dependent measures rather than on effects for any par- ticular measure. No dependent measure alone is critical, but related sets of measures do begin to show consistent patterns which suggest a true finding. In the discussion of results, the focus is on the caregiver data. Results from the child observations are presented in terms of their corroboration or con- tradiction of the results for caregivers. This is not only because the analyses were better able to predict caregiver behavior, but also because the caregiver variables provided a fuller description of the interactions between caregivers and children. Group Composition and its Relationship to Caregiver Behavior As indicated in Chapter Five, group composition occupies a central place in most considerations of the nature of family day care environments. It was thus deemed essential that this study shed some light on the nature of the relation- ship between group size and the family day care routine. Observations provide an ideal way to examine these relationships. Two aspects of group composition in the home were examined: the total number of children present and the ages of the children present. The number of children is a measure of caregiver burden, as well as of potential attention for the individual child. One question was whether an increase in group size was associated with changing patterns of activities and interactions in the home. With respect to age mix, the question was whether particular age mixes in a home were more burdensome than others or required different kinds of care and thereby influenced caregiver and child behavior. *For example, in this study sample, caregiver education and ethnicity are confounded. That is, the Hispanic sample by and large had less education. This confounding means that ethnicity and education often cannot be separated analytically. 85 The analyses showed that both aspects of group composition were strong determinants of the family day care enviornment. In fact, these measures of group composition showed the strongest and most consistent relationship with caregiver and child behavior among all of the independent measures. Total Enrollment and Caregiver Behavior. The caregiver's behavior with children was strongly related to the total number of children in the home. As the number of children in the home increased, interactions of virtually all types between the caregiver and individual children decreased. Thus, for example, the amount of teaching by a caregiver decreases as the number of children in care increases. At the same time, caregivers' interactions with groups of two or more children at a time increased. That is, as group size increased, the internal management of the home had to change; the caregiver focused her attention from individuals to groups. In the process, individual caregiver/child interac- tions were diminished. Total Enrollment and Child Behavior. The group size relationships in the child-focused observations present a complementary pattern. In homes with more children present, children spend less time interacting with the caregiver but more time interacting with other children. Increased group size provides, from the child's point of view, more opportunities for peer interaction. This was true for both toddlers and preschoolers. For toddlers, larger groups were also associated with more time in independent activity and more time monitor- ing the environment (looking around or observing other's activities). Thus, although the caregiver and child observations in each home cannot be con- sidered as two perspectives on the same scene, having been recorded on dif- ferent days, they nevertheless present a consistent story regarding group size. The question of whether increased group size has a negative effect on child behavior is a moot one, since there is no way to evaluate the relative benefits of interaction with peers, interaction with caregivers, and the time spent in in- dependent activity, all of which are considered developmentally appropriate. Further, in larger homes, it is reasonable and probably effective for the caregiver to encourage more interactions among the children, more indepen- dent activities and more group activities. Age Mix and Caregiver Behavior. The strongest patterns for age mix con- cerned caregiver interaction with toddlers. The presence of a preschooler was associated with a significant decrease in all of the caregiver's one-to-one in- teractions with individual toddlers. In effect, preschoolers were able to dif- ferentially draw the caregiver's attention when they were in care with younger children. On the other hand, caregiver interactions with individual preschool children were less strongly influenced by the presence of a toddler. In homes with a toddler, there was a decline in attention for preschoolers only in two types of "noncognitive" interactions with caregivers — play and work. The effect of an infant in the home was seen almost exclusively in a few predictable ways. In homes with an infant present, caregivers tended to display more affection, provide more comfort, offer help more and attend more often to physical needs. Thus, while the presence of an infant had little effect on the 86 kinds or amounts of interaction between caregivers and the individual toddler or preschool child, it did appear to require some special behavior by the caregiver. Further elaboration of the influence of an infant on interaction patterns was found in one of the subsample analyses. Homes were selected where there was at least one toddler and one preschooler present. In this sample, the presence of an infant in the home was associated with less cognitive activity (teaching, language/information and structured fine motor activity) and more helping and attention to physical needs. Thus, in homes where an infants presence means that three age groups are represented, the presence of the infant ap- pears to reduce certain positive kinds of caregiver interaction with older children. It was difficult with our sample to examine the effect on caregiver behavior of school-aged children in the home because observations were conducted in the morning, when most school-aged children were away. Fewer than 30 per- cent of the homes had a school-aged child present during study observations. In this limited sample, the presence of a school-aged child was correlated with fewer cognitive activities for both toddlers and preschoolers and less attention to physical needs. These data should not be interpreted negatively, however, since the presence of a school-aged child in family day care during the school day usually signals some exceptional event such as illness which can be ex- pected to temporarily place greater strain on the caregiver and other children. Age Mix and Child Behavior. The effect of age mix was not as strong for child behavior as for caregiver behavior but, here again, results formed a pat- tern consistent with the pattern for caregivers. For toddlers, the presence of a preschooler was associated with more monitoring of the environment and less interaction with the caregiver — results that parallel the picture presented in the caregiver data. Other results for toddlers were that the presence of a preschooler in the home was associated with more gross motor activity (which was more frequent among preschoolers than among toddlers) and more an- tisocial behavior. As with caregiver behavior, a toddler in the home had little effect on the behavior of preschool children. The presence of an infant had few relationships to toddler behavior, and there were no effects on the behavior of preschool children. Although there were several effects for toddlers, there was no clear pattern. An infant was associated with more affection, more control of another child, and more time alone, and less work and less exploratory fine motor activity with another child. Conversely, the presence of a school-aged child was related only to the behavior of preschool children. For preschoolers, presence of a school-aged child was associated with less language/information activity with the caregiver, less total interaction with the caregiver and less music. One source of additional information about the effects of school-aged children in the home is the SRI morning-afternoon reliability analyses of the observation in- 87 struments.* SRI conducted afternoon observations in a subsample of 12 homes. As school-aged children were present in these homes, the data suggest how caregivers and younger children are affected by school-aged children. The data suggest that the presence of the additional older age group required a shift in the caregiver's attention away from the younger children, especially the preschoolers. SRI found that, compared with the morning observations, in the afternoon caregivers spent significantly more time interacting with school- aged youngsters and less time interacting with preschoolers. Toddlers and preschoolers spent less time watching educational TV and more time watching non-educational TV. Effects for Regulatory Status of the Home The descriptive analyses showed consistent differences in caregiver behavior in sponsored homes compared to regulated and unregulated homes. The regression analyses confirmed these differences. Even with other independent measures accounted for, sponsored homes looked very different with more cognitive teaching activities and less frequent caregiver behavior that did not involve ChH3ren. s, /''Caregiver Behayior. Caregivers in sponsored homes tended to engage more Wte*4n-cognffive activities — teaching, language/information activities and structured fine motor activities. In addition, sponsored homes more often had music activities. Caregivers in sponsored homes tended toward more supervi- sion/preparation and less recreation alone, while the reverse pattern held in unregulated homes. The pattern of effects for regulatory status suggests that sponsored homes look more center-like and less home-like. The reasons sponsored homes look different are probably multiple and complex. An important difference in- volves the caregivers themselves, who in sponsored homes more often had received some training. It is also likely that caregivers who have become af- filiated with a sponsor have a different orientation toward their work than caregivers operating alone in that they are more inclined to preceive themselves as professionals. As an example, an attitude questionnaire was administered to caregivers in which they expressed their philosophies about child-rearing, education and related topics. In this interview, sponsored caregivers more often stressed the importance of the education rather than the social environ- ment. Such differences might explain the more structured environment in sponsored homes. During the analyses, another possible explanation for the pattern of ac- tivities that distinguished sponsored homes was raised. It was pointed out that in the study sample, enrollment and regulatory status of the home are con- founded; sponsored homes tended to be larger, which might account for a higher frequency of structured activities. To test this hypothesis an analysis *See Volume IV of the National Day Care Home Study Report, The Observation Component by SRI International, for further detail on this analysis. 88 was conducted including only homes enrolling three or fewer children, thus eliminating the difference in size between sponsored and nonsponsored homes. In this sample, not only did the same differences hold, but the results were stronger and involved a wider set of behaviors. In this special sample, the sponsored homes also tended to exhibit significantly more caregiver conversa- tion with preschoolers, more helping and less housekeeping. Thus, it is possi- ble to conclude that there are marked behavioral differences between spon- sored and nonsponsored providers, and these differences are not attributable to the size of the sponsored homes. Child Behavior. The effects of regulatory status on behavior were less strong and less systematic for the child variables. Results that were consistent with those for caregivers included the following: first, for both toddlers and preschool children, there was more structured fine motor activity in sponsored homes; and second, for toddlers, there was less interaction with the caregiver in regulated homes. The remaining effects for children were too scattered to interpret. Effects for Caregiver Experience, Education and Training The effects of experience, education and training were carefully analyzed, as these caregiver characteristics are often used as surrogate measures of caregiver competence. Experience was quickly eliminated, based on the available data, because there were too few significant relationships between ex- perience and caregiver behavior and this variable consequently did not distinguish caregivers from each other on the behavioral dimensions. The remaining two variables required very careful analysis because each was strongly- related to ^a-variable used to design the study. Education was related to caregiver ethnicityj^Hispanic caregivers had by far the lowest levels of educational attainment. Thus analyses of the observation data taken as a whole cannot clearly distinguish between the effects of education and those of ethnicity. To estimate the separate effects of caregiver education, analyses had to be conducted for each ethnic group separately, to see if the effects of educa- tion would be consistent for each group. In a similar vein, whether or not a caregiver has received training is related to whether or not she is sponsored. Most sponsored providers are trained, but most nonsponsored providers are not. Thus, to determine whether training itself makes a difference, training must be analyzed separately for sponsored, regulated and unregulated caregivers. It is only by showing that training makes a difference for each of these groups that we can clearly conclude that training is beneficial. §£^ Years of Education and Caregiver Behavior. Overall, more education was associated with more teaching, more language/information and prosocial ac- tivity, and less helping, directing, household work, attention to physical needs, and positive affect. Most of these relationships could be attributed to caregiver ethnicity as well, as nearly every measure of caregiver behavior related to education was also related to caregiver ethnicity. In order to help disentangle 89 to f / the effects of education and ethnicity, analyses were conducted for each ethnic group separately. The question was whether the same effects for education would be found for each ethnic group. Unfortunately, these analyses were not very helpful in separating out the effects of education and ethnicity, as the relationships between education and caregiver behavior varied across the ethnic groups. Education had little effect in the sample of Black caregivers; it had the largest and most interpretable effects in the sample of Hispanic caregivers; the effects in the sample of White caregivers were scattered. Specifically, among Hispanic caregivers, education was associated with more teaching, both cognitive and social, and less non-educational activity. Education was related to more teaching, play/participation, language/in- formation and prosocial activities; it was related to less conversation with adults, household work with preschoolers, directing preschoolers, and atten- tion to physical needs. The larger effects for Hispanic caregivers, compared to White and Black caregivers, may have occurred because a wider range of education is represented in the Hispanic sample. The effects among White caregivers are consistent in tone with the results for Hispanic providers, but fewer in number. For White caregivers education was related to less control, household work, directing, and exploratory fine motor activities. Even the two significant effects among Black caregivers were consistent with the above find- ings — more education was associated with less directing and less work with preschoolers. Consistent results in the above analyses permit the limited conclusion that caregivers with more education are less likely to engage in household chores and less likely to direct children. Among Hispanic caregivers (or, perhaps, in any sample with a more extensive range of education, particularly at the lower end of the scale), higher education is also associated with more cognitive ac- tivities and encouragement of prosocial behavior. Thus, while education does appear to have some consistent relationship with caregiver behavior across ethnic groups, these relationships do not appear to be significant contributions to the family day care environment. Among Hispanics, it is quite possible that education makes a difference, but alternative hypotheses cannot be ruled out. Year of Education and Child Behavior. The analyses of the child variables did not provide much support for any conclusions about caregiver education and ethnicity, as there were very few significant effects and those that were found were consistent with the caregiver findings. For toddlers, more educa- tion of caregiver was associated with less control. For preschool children, higher education was associated with more structured fine motor activity. Training and Caregiver Behavior. Although the effects of education on caregiver behavior are uncertain, the effects of training are strong and positive. This is perhaps the most encouraging of the study findings, for it means that an investment in caregiver training can influence the ways in which caregivers interact with children. Thus, quality of the family day care environ- ment can be enhanced by this means. 90 To disentangle the effects of training and sponsorship, effects for training were examined separately in sponsored homes, regulated homes and unregulated homes. In sponsored homes, nearly 80 percent of the caregivers had some training, versus 30 percent in regulated homes and only 20 percent in unregulated homes. The question asked in this analysis was whether trained caregivers behaved differently from untrained caregivers, when regulatory status was taken into account. Analyses showed that there were consistent positive effects for training in each group of caregivers (see Table 23). The trained caregivers in regulated and sponsored homes looked similar. In these groups, training was associated with more teaching, helping and dramatic play and with less activity that did not in- volve interaction with children. In sponsored homes, caregivers who had been trained also exhibited more structured teaching— more language/information and structured fine motor activities. In both regulated and unregulated homes, training was associated with more comforting and less time away from the children. Tabi(e 23. — Caregiver Behaviors Associated with Training.^) Sponsored Regulated Unregulated More Frequent Behaviors. Language/Inform Dramatic Play Structured Fine Teach Motor Help Music Comfort Dramatic Play Teach Help Comfort Positive Affect Less Frequent Behaviors Recreation Alone Supervise Supervise Recreation Alone Training and Child Behavior. There were few significant relationships be- tween caregiver training and children's behavior. However, all of these rela- tionships were consistent with the findings for caregivers. Toddlers in homes with trained caregivers more often engaged in structured fine motor activity, conversed less often with other children and had to be controlled less often by the caregiver. For preschoolers, caregiver training was associated only with more music/dramatic activities. Effects for Related Children in the Home A final set of analyses was conducted to see if caregivers behaved differently in homes that provided care for the caregiver's own child or a related child than in homes that provided care only to nonrelatives. Earlier chapters characterized homes with the caregiver's own child in care and those with a relative's child in care. Care of one's own child tended to occur in the homes of young, well-educated, White caregivers in unregulated homes. Nonresident relative care also tended to be provided by unregulated caregivers, but by 91 older, less educated, non- White caregivers. If either type of home was shown to look different from other homes, it could be caused by a number of characteristics of the home in addition to the type of child. In fact, for both types of home, the pattern of effects seemed as much attributable to the caregiver's situation as to the presence of a related child. Care for One's Own Child. In general, homes where the caregiver's own child was present had fewer child-centered activities and less structured activi- ty. More specifically, caregivers in these homes tended to engage in more housekeeping and control of antisocial behavior and less teaching, conversa- tion with toddlers, play/participation and TV. In general, there was less in- teraction between the caregiver and toddlers when the caregiver's own child was present. Looking at the child observation variables, the only noticeable ef- fects were for preschoolers, and these effects were consistent with the caregiver findings. Presence of the caregiver's own child was associated with more time spent with other children, more exploratory fine motor behavior with other children, more music activities and more time reading books. One interpreta- tion of these results is that homes where the caregiver's own child is present look more ''home-like" — that is, the caregiver continues her own activities and less often initiates structured activities with the children. These homes also tended to be smaller, and this pattern may be less feasible in larger homes. To disentangle the effects of the caregiver's own child in the home and the background characteristics of these caregivers, the presence of the caregiver's own child was examined within the sample of White unregulated caregivers. (Sixty percent of this group cared for their own children). Even in this subsam- ple of caregivers who shared a pattern of background characteristics, the ef- fects of caring for their own children were significant. Caregivers who had their own child at home engaged in less teaching, conversation and play with toddlers, and fewer language/information activities; they also engaged in more housekeeping and control of antisocial behavior. Relative Care. Caregivers providing care to a relative's child (or children), like those who had their own child at home, tended to exhibit less cognitive teaching. These caregivers also showed less play/participation and helping, all of which suggest less interaction with children. At the same time, caregivers providing care to a relative's child exhibit more directing and do more household work. Comparable effects were found on the child variables. That is, for toddlers, relative care was associated with less language/information and less educational TV. For preschool children, it was associated with less language/information, less structured fine motor activity, less attention- seeking, more antisocial behavior with other children and more prosocial behavior. Conclusions The observation system developed for the NDCHS proved to be very sen- sitive to the home process. It permitted the coding of noteworthy child and caregiver behaviors and permitted us to discern the effects on these behaviors 92 of variables of policy and programmatic significance such as the caregiver's training, enrollment, age mix and regulatory status. In general, the observations showed family day care homes to be positive en- vironments for children. It was observed that caregivers spend a considerable portion of their time in direct interaction with children, and the time spent with children seems to be appropriate to the needs of children of various ages. Caregivers rarely expressed any negative affect toward the children. The caregivers' homes were generally safe, home-like environments which were less structured and homogeneous with respect to children's ages than day care centers. Some of the most interesting implications of the study findings are based on comparisons among different types of family day care homes and among caregivers with differing degrees of preparation for child care. Specific ex- amples of important differences found in such comparisons include the following: • On the whole, the types of activities in unregulated and regulated homes were similar to one another, but sponsored homes showed a dif- ferent pattern, placing more emphasis on cognitive and expressive ac- tivities. These homes were more suggestive of a preschool environment. The differences in caregiver activities were generally small except for teaching behaviors. Teaching occurred almost 50 percent more often in sponsored homes, where it accounts for 17 percent of the caregiver's time, compared to 12.1 percent and 12.8 percent in unregulated and regulaTe^TfToTires^j^spectively. • Caregiver trainingjwas also found to influence the pattern of activities in the family day care home. Caregivers who had some child care train- ing tended to display more teaching, language/information activity, music/dramatic play, and comforting. This pattern of behaviors sug- gested more structured teaching on the part of trained caregivers. The patterns associated with training are very similar to those associated with sponsored homes, referred to above. This remained true even after training and regulatory status were unconfounded. • Caregiver and child behaviors also tended to vary across homes with different group sizes and age mixes. As the number of children in the home increased, interactions of virtually all types between the caregiver and individual children decreased (with the exception of control ac- tivities). At the same time, caregivers' interactions with two or more children increased. Child-focused data also complemented the pattern of caregiver behavior which was observed: in homes with more children present, children spent less time interacting with the caregiver but more time interacting with other children. • The presence of the caregiver's own children or of a relative in care ap- pears to affect the caregiver's behavior towards the children in care. In general, the homes where the caregiver's own child was present could be characterized as less formal, with more activities that were not 93 centered on the children and more activities that were less structured. This also tended to be true for caregivers who cared for a nonresident relative. These observation data, while strongly suggestive of differences in caregiv- ing style, should not in themselves be considered as the basis for evaluating the quality of care in different types oLfamily day care homes. For example, the long-term effects on children of different emphases are not well established. It will be recalled that even though differences in the amount of time different groups of caregivers spent in particular kinds of activities were observed, all caregivers devoted a substantial portion of time to their overall involvement with the children. The patterns of differences that were found should, however, prove useful both in establishing guidelines for parents seeking specific activity patterns for their children, and in informing policymakers of the probable effects of alternative family day care policies on home processes. 94 Chapter 8 COSTS OF FAMILY DAY CARE* Substitute child care is a necessity when mothers work. While the suitability of the arrangement to the child is of considerable concern to parents in making a selection from among the various day care alternatives, there are other fac- tors that will frequently determine which child care arrangement will be selected. One of the most important of such factors is the cost of care. Often a mother will relate the cost of child care to the salary she herself earns rather than to the total family income. Thus, if it costs almost as much to keep a child in care as a woman can earn outside of her home, one will often hear her say that it does not pay for her to work. Thus women's salaries in the marketplace set an effective cap on the costs of child care. Since women's salaries tend to be relatively low, it is not surprising that fees for child care have been correspondingly low. What may be surprising to some is that, even where day care is subsidized by the federal or state government, the resultant caregiver wage remains substantially below the minimum wage. The Cost of Family Day Care to Parents or to the Government Nearly all parents of children in regulated or unregulated homes absorb the total cost of child care; most children in sponsored care are subsidized entirely or in part by the government through Title XX of the Social Security Act, the USDA Child Care Food Program or state programs. In this section, we do not differentiate among parents and the government as purchasers of care, but rather have combined these two groups to contrast the point of view of those who pay for care with that of those who are paid to provide care.** Most parents pay a certain hourly fee to keep a child in care. As indicated in Table 24, the average hourly fee paid to a caregiver in NDCHS homes was $0.59,*** but the range in fees was quite large. The very lowest average fee for any type of home in the study sample was $.44 per hour for unregulated Black homes. The maximum average fee for any type of home was $1 .00 per hour in •Information in this Chapter is abstracted from the NDCHS Final Report, Volume II, The Research Report, prepared by Abt Associates. **The point of view of parents is presented in Volume IV of the NDCHS final report series, Volume IV, Parent Study Component Data Analysis Report , by CSPD. ***These data were gathered in 1977 and 1978. Because of inflation, it is expected that costs in all categories have risen in the intervening years, so that these numbers underestimate current costs. 95 White sponsored homes. As is indicated in Table 25, the average weekly fee per child paid to a caregiver was $20.85. Weekly fees ranged from $16.54 for unregulated Hispanic homes to $31 .80 for White sponsored homes.* The information in Tables 24 and 25 indicates that the fees paid per child vary considerably across regulatory status and ethnicity. Sponsored and regulated homes tended to have higher mean fees than unregulated homes. Similarly, White caregivers had higher fees than Black and Hispanic caregivers. A rough rule of thumb is that the more regulated the home, the higher the per-hour fee, and the higher the average income of the community which the home serves, the higher the per-hour fee. Table 24.— Mean Hourly Fee Per Child. a Sponsored Regulated Unregulated White Black Hispanic 1.00 0.62 0.58 0.68 0.56 0.58 0.68 0.44 0.45 .70 .52 .51 0.70 0.63 0.54 .59 a This table includes children who are cared for without charge. Table 25 .— Mean Weekly Fee Per Child. a Sponsored Regulated Unregulated White Black Hispanic 31.80 24.68 24.49 23.68 21.61 21.42 19.70 16.57 16.54 22.54 19.78 19.37 26.36 22.65 17.80 20.85 a This table does not include children who are cared for without charge. To understand the economics of day care, it is important to compare the fees charged for family day care with those charged for center care. The Na- tional Day Care Study found that fees for center care vary from $17 to $27 per week for parent fees and from $25 to $32 per week for government subsidized day care slots. 1 Comparing the parent fee figures with the weekly rates for regulated and unregulated family day care, and comparing the government rates in centers with the weekly rates for sponsored family day care, it is evi- dent that though center care may be slightly more expensive the differences are not great. As we shall point out below, the major differences are on caregiver earnings. In general, as the number of hours a child is in care increases, the per-hour fee decreases; part-time fees were significantly higher than full-time fees across all sites. Across all children, the average hourly fee for a child in part-time care was $0.83 and for a child in full-time care, $0.54. This rate does not differ * Because many families need day care for more than one child, the weekly day care expenditures of the family may be greater than the weekly costs for a single child. Interviews with parents showed that the median weekly cost to the family was $26.35. Fourteen percent of parents paid less than $ 1 5 per week, 29 percent between $ 1 5 and $25 and 57 percent over $25 per week. 96 much by age of child; for example, the rate for preschool children is similar to that for school-aged children. Fees charged to relatives of the caregiver are also lower than those charged to nonrelatives. One reason for the lower rate in homes with at least one relative is that free care is frequently provided for relatives. Of the 201 homes in the sample with at least one relative in care, 1 percent (42 homes) had at least one nonpaying child. In homes without relatives, only 3 percent (15 homes) had one or more nonpaying children. Thus free care is more common in unregulated homes, but is fairly evenly distributed across sites and across caregivers of different ethnicities. When nonpaid relative care is subtracted from the sample, however, caregivers still charge less for relative care than for the care of nonrelatives (see Table 26). Overall, the average fee in homes with a relative in paid care is $0.47; in homes with no relative, it is $0.64. The most substantial differences appear in unregulated care, where there is a 45 percent increase in fee from relative to nonrelative care ($0.42 versus $0.61). Fees may also vary for families with more than one child in care. Specifi- cally, in San Antonio and Philadelphia caregivers were asked if they charged more, the same amount, or less for siblings in care. Two percent (five caregivers) said they charged more for siblings, 65 percent reported their charges were the same and 33 percent said they reduced the fee when siblings came together for care. Table 26. — Mean Hourly Fee Per Child by Whether or Not a Relative is in Care. a Sponsored Regulated Unregulated Relative No Relative Relative No Relative Relative No Relative White Black Hispanic 1.09 0.55 0.62 0.98 0.63 0.57 0.65 0.54 0.45 0.68 0.57 0.60 0.50 0.35 0.44 0.72 0.52 0.47 0.67 0.70 0.56 0.64 0.42 0.61 a This table does not include free care. Caregivers' Income and Costs In this section our point of view shifts from that of the parent to that of the caregiver, as we look at the wages the caregiver earns and the costs which she incurs in running a family day care home. We shall show that, though some parents may find child care to be an expensive service, from the caregiver's perspective child care income is very low. Caregiver Wages Table 27 displays the average hourly wage of caregivers in the sample, counting all the children in care. The average hourly wage of the 666 caregivers represented in this table is $1.25, although the range is from -$0.58 (where the caregiver is actually losing money) to $7.84. In general, caregivers in spon- 97 sored homes tended to earn higher wages than those in regulated homes, who in turn earned somewhat higher wages than those in unregulated homes. White caregivers earned higher wages than Black or Hispanic caregivers. Table 27 . —Mean Net Hourly Wage. a Sponsored Regulated Unregulated White Black Hispanic 2.28 1.82 1.65 1.66 1.17 1.04 1.12 0.80 0.53 1.53 1.15 0.90 1.92 1.39 0.85 1.25 a Hourly wage is calculated as weekly revenue from fees minus weekly costs for food, supplies and insurance divided by the number of hours of care provided weekly. Where the cost of food, supplies or insurance was not known, zero cost was assumed. Histogram of Table 27 Mean Net Hourly Wage WAGE/HR SPON REG UNREG To view caregiver wages in perspective, we must consider the relationship of their wage rates to wage rates in the population at large. Figure 7 graphically il- lustrates the distribution of hourly wage rates for the caregivers in this sample and pinpoints significant wage rates along a continuum. For example, the minimum hourly wage in 1977 was $2.30, almost twice the average wage of family day care providers. The 1977 Poverty Line was defined for wages of $2.88 per hour and the Low Income budget line was set at $4.81 per hour. Relatively few caregivers reach either of these amounts in their earnings from day care. Eighty-seven percent earn wages below the minimum wage; 94 per- cent have earnings below the Poverty Line and a full 99 percent are below the Low Income line. Thus, although child care may be a significant cost for parents, most pro- viders earn only a scant income from their jobs. Most are earning wages that would keep them significantly below the poverty line if child care were their only source of income. 98 The median household income of day care home providers is $10,500. The income they receive from day care is included in this figure. This implies that most providers have other sources of household income besides child care. However, even with two or more sources of income taken together, caregivers' families live only slightly above the Low Income cutoff set by the Department of Labor (shown in Figure 7). For families with more than one breadwinner, the operation of a day care home appears to make the difference between the Poverty Level income and a Low Income budget, but does not push their total income level far above this Low Income line. On the other hand, those caregivers whose sole or major source of income is family day care subsist substantially below the poverty line. Figure 7 Distribution of Caregiver Wage Rates p e Average Wage for Family Day Care Providers $2,600/yr 1977 Minimum Wage: $4,7807 yr 1977 Poverty Line: $6,000/yr 1977 Department of Labor Low Income Budget $10.000/yr $1.25 $2.30 $2.88 $4.81 r c e n t 30 j— 29% a g e 24% f 20 r C 15% a 13% r e g j 10 I — 9% V e 4% 3% r 1% ^ 1% s I 1 I I S0.00 $1.00 $2.00 $3.00 $4.00 $5.00 and $0.50 $1.50 $2.50 $3.50 $4.50 over He »urly W age Ri ite 99 Caregiver Revenue Caregiver revenue per week* offers the best approximation of the provider's stable income over time. Across all groups of homes, the average weekly revenue from child care is $73.92 (see Table 28). If this amount represented net income, it would mean that caregivers earned an annual salary of $3,844. Table 28 presents a breakdown of average weekly revenue by ethnicity, regulatory status and whether or not the caregiver provides relative care. This table illustrates over again the large discrepancy between the various groups of providers in average weekly earnings. Table 28 .— Mean Weekly Revenue from Fees by Type of Care. a Sponsored Regulated Unregulated Total Relative No Relative Relative No Relative Relative No Relative White Black Hispanic . . 128.70 123.36 82.95 123.60 130.07 93.99 86.13 79.83 65.35 96.80 79.52 72.50 56.50 40.41 28.74 59.61 64.60 30.56 84.00 78.07 52.55 Total . 108.26 117.15 J 78.95 87.60 86.31 38.47 54.37 48.23 V 115.97 73.92 a This table includes children who are cared for without charge. To reach the net income or wages of caregivers, costs associated with such items as food and supplies must be subtracted from this amount. Caregivers thus actually earn much less than it appears at first glance. Supplies consist of any nondurable equipment and materials such as clean- ing materials, toothpaste, paper products and most toys. The weekly cost of such supplies averaged $1.55 per week per child, and it appears that such costs are of little importance in the total cost of care. Food is the caregiver's major out-of-pocket cost. Table 29 shows caregivers' average per-child weekly food expenditures. Across all categories, the mean was $5.67 per child per week. Sponsored homes incurred higher food costs than regulated or unregulated homes. Black and Hispanic caregivers both spent more on food than did White caregivers. Because of these high food costs with respect to the caregiver's overall in- come, there has recently been substantial caregiver interest in participation in the Child Care Food Program, run by the Department of Agriculture. Because program regulations restrict participation to homes which are part of non- profit family day care systems, there has been a remarkable growth in such systems in the past few years.** ♦Weekly revenue was computed by multiplying weekly enrollment by the mean weekly fee charged per child. ** A discussion of the Child Care Food Program is contained in the NDCHS Research Report. 100 Table 29 . —Mean Weekly Expenditure per Child for Food. Sponsored Regulated Unregulated White 5.46 4.78 4.43 4.72 (24)" (81) (77) (182) Black 8.38 6.19 6.18 6.79 (40) (49) (56) (145) Hispanic 4.13 5.69 6.05 5.78 (7) (35) (52) (94) 6.97 5.39 5.42 5.67 (71) (165) (185) (421) a Number of homes. In order to determine net income from caregiving, the costs of food, sup- plies and other costs such as day care insurance were subtracted from the revenue a caregiver received from parent fees. These computations are sum- marized in Table 30. Two means are presented for each of the nine cells, a minimum and a maximum for that cell. Two calculations were necessary because of the large number of homes for which data were missing for one or more of the cost variables. The maximum estimate of net income was calculated by assuming that the missing costs were zero; this clearly overestimates the net income. The minimum estimate was calculated by taking the average weekly revenue in a cell and substracting the average costs of food, supplies and insurance of homes within that cell. Such a computation has the effect of creating a reasonable estimate of net income which, if anything, is lower than the actual net income. Table 30.— Average Weekly Net Income per Home from Caregiving. Sponsored Regulated Unregulated White Black Hispanic 103.54^ 69.75 44.75 63.78 1 1 1 .80 b 82.26 51.05 73.14 90.83 49.19 32.87 48.93 108.22 62.25 42.00 62.26 76.00 35.91 16.41 30.85 86.79 52.29 24.01 43.37 86.11 57.11 32.16 50.27 102.84 70.90 39.91 62.09 a Minimum estimate of mean net income. This was calculated from average weekly revenue minus the mean weekly costs of food, supplies and insurance. b Maximum estimate of mean net income. This was computed by assuming the costs of insurance, food and supplies to be zero when data were missing. The average weekly net income averaged across all 723 homes in the sample was thus within the range of $50.27 to $62.09. Sponsored homes had somewhat higher incomes than regulated and unregulated homes; White caregivers earned higher incomes than Black or Hispanic caregivers. And, once again, if we translate these weekly figures into yearly income figures, caregivers can be seen to earn an average of $2614 to $3229, substantially below the Poverty Level. 101 Histogram of Table 30 Average Weekly Net Income per Home from Caregiving INCOME RANGE SPON REG UNREG These weekly incomes can be compared to the weekly income of classroom staff in day care centers. Child care fees for parent or the government are ap- proximately the same for family day care as for center care. However, center wages are significantly higher than family day care wages. Across all centers the lowest average weekly wage reported in the National Day Care Study was $89 per week and the highest was $190 per week. 2 Note that only the income of sponsored White and Black family day care providers even fall within this range, with a maximum weekly income of $111.80. All other family day care providers earn less than the lowest paid center classroom staff, most substan- tially less. The reason for this wage differential is easy to understand and fundamental to the nature of the difference between family day care and center care. The amount a caregiver can earn is directly tied to the number of children for whom she cares. Across all centers the average child/staff ratio reported was 6.8 with the number going as high as 7.9 in one profit-making category. In family day care, on the other hand, enrollments averaged 4.3 per home in sponsored care, 4.0 in regulated care and only 2.8 in unregulated care. Thus, although the fee per child is about the same, the only way family day care pro- viders can earn approximately as much as center classroom staff is to care for as many children as regulations allow. This is more than the standard policy of most sponsors permits, while for regulated and unregulated caregivers it is typically more children then they can enroll or wish to care for. Further "Costs" to Caregivers Family day care providers have a long work day. Children are typically dropped off by parents on their way to work and picked up some time in the 102 early evening on their way home. An individual child thus spends with the caregiver not only the portion of the day the parent is at work, but also the portion of the day that the parent spends commuting to and from work. When added on to the typical 30- to 40-hour work week of most parents, this pro- duces a caregiver work week well in excess of 40 hours. The caregiver work week is likewise affected by staggered schedules of the day care children. It is the rare home in which all children come and go at the same time; on the contrary, arrival and departure times are generally arranged for the convenience of the working parent. Thus, some children may arrive as early as 7:00 or 8:00 and leave earlier in the day, and others may not arrive un- til 8:30 or 9:00 but stay considerably after 6:00. This uneven scheduling similarly lengthens a caregiver's work week considerably beyond the typical work week. These two factors explain why the vast majority of caregivers interviewed in the NDCHS work long days, producing a total work week of 40 to 60 hours; across sites, the average work week was 50 hours long. An additional 10 per- cent of the caregivers provided care in excess of 12 hours a day, and some even provided overnight or weekend care as part of their regular service, producing a work week in excess of 100 hours. The remaining 15 percent serve children for fewer than 40 hours a week, and, in our entire sample, only 7 percent care for children for fewer than 30 hours a week. The provision of family day care services is obviously a full-time job in the truest sense of the term.* Providers were asked whether the length of their work day was a problem to them. Given their long hours, it is not surprising that only a handful would care to increase their work day. Most would not like to decrease their work day either; only about one-quarter of those interviewed would like to curtail their hours. In essence, then, most family day care providers view the long hours and minimal time off as part and parcel of their job and do not express much dissatisfaction with this routine. Conclusions From the parent's perspective, family day care may appear a costly endeavor. An average of 60 cents per hour must be paid for the care of each child. Since many children are in care for 40 to 50 hours a week, this expense could easily exceed $30 per week. From the provider's perspective, however, family day care is not a lucrative profession. The average weekly wage for pro- viding care is $50.27 to $62.09 after payments are made for food, supplies and insurance. As a result, many caregivers' earnings are significantly below the poverty level. They work long hours, frequently have no provisions for sick time or for vacations, and often are not even aware of tax advantages for which they are eligible. *No differences in the caregiver's work week were noted across sites, ethnicity or the regulatory status of the home; in all these domains, the average caregiver works a 10-hour day. 103 One central policy question with regard to the cost of care is the impact on weekly fees of the extension of minimum wage requirements to family day care providers. Would this stretch the ability of parents (or public funding sources) to pay beyond the breaking point, or would it help the population of family day care providers without seriously hurting parents? Table 31 displays the impact on weekly fees of the simulated extension of minimum wage requirements to family day care providers. Two hypothetical wage levels are compared with current fee structures which appear in the first column. Based on the 1978 federal minimum wage of $2.65 per hour, for ex- ample, the figures in the second column give the mean amounts that would have been charged per week for each child in care to assure the provider a gross hourly wage of this amount. Fees in sponsored homes would have to rise an average of 28 percent to cover these wages. In regulated homes they would have to rise an average of 59 percent and in unregulated homes an astounding 190 percent increase would be required.* In general, the size of the increase is inversely related to current gross hourly wages. Clearly, the size of the required increase is more than enough to threaten most parents' ability to pay. Therefore, although providers earn little from their caregiving, to permit or mandate improvements in their earning power would impose severe burdens on consumers of family day care — both parents and the government — a dilem- ma with no clear solution. Table 31 . — Mean Weekly Fee Per Paying Child Under Current and Alternative Minimum Wage Requirements for Providers. If Minimum Wage Were: Fee Currently $2.65/hr. $2.90/hr. Charged (as in 1978) (as in 1979) (no minimum Percent Percent wage ^ Fee Increase Fee Increase Sponsored: White $29.61 $33.73 a 14 $36.92 25 Black 25.41 31.82 25 34.83 37 Hispanic 24.49 38.56 57 42.20 72 Average 26.50 33.89 28 37.09 40 Regulated: White 23.68 33.35 41 36.50 54 Black 21.61 37.76 75 41.32 91 Hispanic 21.42 43.12 101 47.19 120 Average 22.65 36.08 59 39.48 74 Unregulated: White 19.70 47.02 139 51.45 161 Black 16.57 44.57 169 48.78 194 Hispanic 16.54 69.40 320 75.95 359 Average 17.80 51.68 190 56.56 218 a Each fee was computed as [minimum wage * 40 hours at regular pay] plus [minimum wage * 1.5 x 10 hours for overtime pay] divided by the mean number of children enrolled. *The smaller the average group size cared for in the home, the larger the per-child increase in fees would have to be in order to produce minimum wage for the caregiver. 104 Histogram of Table 31 Mean Weekly Fee per Paying Child Under Current and Alternative Minimum Wage Requirements for Providers FEE 70 $2.90/HR $2.65/HR CURRENT FEE SPON REG UNREG Coelen, C, F. Glantz and D. Associates Inc., 1978, p. 60. Ibid., p. 65. Notes Calore, Day Care Centers in the United States 1976-1977, Abt 105 Chapter 9 FAMILY DAY CARE SYSTEMS* Family day care systems have developed as an alternative to center care, par- ticularly for the increasing numbers of infants and toddlers needing care, but also for preschoolers and school-aged children. They arrange for day care through a network of affiliated family day care homes and thus provide a necessary framework** for the provision of child care subsidies and other ser- vices to children and families within a family day care context. While such an organization is not explicitly required in order to receive child care subsidies under Title XX of the Social Security Act, it is a requirement for the USDA Child Care Food Program. That program requires that family day care homes have nonprofit sponsorship in order to be eligible for payments for food pro- vided to the day care children. The Child Care Food Program has thus pro- vided the strongest impetus for system formation in recent years. We estimate that there are now over 30,000 family day care homes nation- ally under the sponsorship of such umbrella organizations. Although spon- sored homes represent but a small portion of the total number of family day care homes (including unregulated homes), they are important beyond their numbers, primarily because they provide care for most state and federally sub- sidized children in family day care settings. Eighty-five to 90 percent of system slots are subsidized, whereas relatively few slots in nonsponsored homes are publicly funded. Family day care systems frequently relieve welfare and regional human ser- vice departments of many administrative tasks necessary to the delivery of sub- sidized care through family day care homes. Systems often determine family eligibility for subsidized child care and determine if family income level and circumstances warrant free or reduced-fee care. The system subsequently determines what fee, if any, is to be paid by the parent and bills the govern- ment for the remainder. In addition to handling these financial arrangements, most systems also select and train their own caregivers as well as providing a range of services to families, including medical and dental screening, emer- gency care, nutritional assistance and referral to other community organiza- tions. •Further information on family day care systems is found in Volume V, Family Day Care Systems Report, by Abt Associates. **Some state agencies act as sponsoring agencies for family day care homes, making these homes eligible for a variety of day care subsidies. 107 This chapter describes the 22 family day care systems studied during the NDCHS. It begins by summarizing the characteristics of these day care pro- grams — the services they offer, their providers, directors and children. The chapter concludes with a discussion of program cost characteristics and finan- cial issues in sponsored family day care. Profile of Family Day Care Systems Table 32 summarizes characteristics of the 22 family day care systems whose directors were interviewed as part of the NDCHS. Nine of these systems were in Los Angeles, eight in Philadelphia, two in Texas, one in Arkansas and two in the Greater Boston area. These systems ranged in age from one to 27 years; median age was approximately 5 years. Because a recent California state law* had promoted the start-up of family day care systems, Los Angeles programs were considerably younger (median age 8.8 years). Many systems in these other sites had their origins in established religious organizations that have long provided foster care, adoption services and family day care through a variety of income sources. In the past, lack of start-up funds has limited the formation of new family day care systems to existing private social service agencies which can support a new program financially until the system can generate income on its own. However, with the advent of the Child Care Food Program, some of the finan- cial impediments to establishing systems have been removed, and more systems are now starting without a pre-existing social service base. Further- more, in some instances where private social service agencies have been unwill- ing or unable to act as a family day care sponsor, public social service agencies are themselves performing this role. Systems included in the NDCHS varied widely on numerous program dimensions. The smallest had only 4 providers, whereas the largest had 135.** The smallest enrolled only 16 children, and the largest enrolled 421. The number of enrolled children per home also varied from fewer than two children per home to almost seven. The great majority of systems, however, assign four or fewer children per home, substantially fewer than federal day care requirements allow. In those systems where more children are assigned per home, system directors generally (but not always) compensate for the ad- ded caregiver burden by providing the caregivers with helpers and/or giving them additional training. In fact, the system director is the pivotal figure in the system's functioning. Ordinarily she is the one who sets the program's tone. However, her philosophy and style are a function of such factors as her experience and education, factors which vary substantially from director to director. Direc- *In California, Assembly Bill 3059 encouraged the start-up of family day care systems during Fiscal Year 1976-1977. ** Although a 135-provider system is considered a large family day care system, systems do range in size to well over 1000 providers. 108 Table 32.— Profile of 22 Family Day Care Systems. Number of Systems Number of Systems Age of Program < 4 years 8 4-7 years 10 ^ 7 years _4_ 22 Median = 5 Range = 2-27 Number of Providers: <16 10 16-30 7 31-45 1 46-60 1 106-120 J_ 22 Median = 16 Range =4-135 Number of Children Up to 50 10 51 to 100 7 101 to 150 2 151 to 200 1 over 200 _2_ 22 Median = 46 Range = 16-421 Number of System Children per Home a : 1.0 to 2.0 4 2.1 to 3.0 6 3.1to4.0 7 4.1 to 5.0 4 5.1 to 6.0 6.1 to 7.0 J_ 22 Median Range 3.5 1.8-6.8 Affiliations/A uspices: Religious organizations 5 Community/voluntary organizations 10 Mental health associations 2 City government 1 University 1 Child care organizations 3 (Head Start 1) (Center 2) 22 Child Care Food Program: Participant 14 Nonparticipant _8_ 22 a No data were collected on nonsystem children enrolled by providers without exclusive use agreements; the presence of such children is therefore not reflected in this table. tors' years of paid experience in a child care-related field ranged all the way from 2 to 30 years (with a median of 1 1 years). Their years of formal education ranged from only two years of college to the equivalent of a master's degree. Directors' salaries mirrored this variation in background, ranging from $5,428* to $26,400. The director is often in charge of the critical area of fund raising. Study data show a high correlation between the director's experience in child care and the amount of noncash contributions to the system as a percentage of the system's overall operating costs. That is, the more experienced the director, the better able she** is to secure the supplementary resources needed to support the system's operations. These resources, as we shall show below, are often used to provide additional services both to the families served by the system and to 'The lowest salary represents a part-time directorship. '*A11 system directors in the NDCHS were women. 109 the children in care. These in-kind contributions (primarily labor) either go directly to providing such services as medical screening from local clinics, or they relieve program staff from minor daily tasks, freeing them to provide other services. The Relationship between Sponsoring Agencies and Their Affiliated Providers The unique relationship between providers and sponsors has developed primarily in response to issues of wage and benefit compensation. Family day care providers have been and continue to be one of the lowest income groups of workers in the U.S. One of the major cost issues underlying the present pay- ment level for family day care is the trade-off between wages of providers on the one hand and limited public dollars and parents' ability to pay for child care on the other. This inherent conflict, coupled with state regulation of minimum wages, has induced most family day care systems to utilize personal services contracts with providers in order to avoid the more costly minimum hourly wage, which many systems feel they cannot afford. Laws on unemploy- ment compensation have influenced systems in a similar fashion. If providers were employed directly by systems and paid hourly wages, programs would be forced to pay minimum wages and contribute to unemployment insurance, workmen's compensation and social security taxes. In addition, they would deduct local, state and federal taxes from providers' earnings. Systems might also be forced to pay overtime increments for providers working more than 8 hours daily or 40 hours weekly (an almost universal occurrence). As a result of these threatened increased costs from regulatory forces, all but one of the systems subcontract with providers for their services rather than treating providers as employees. Most have written contracts or oral agreements that define the relationship between the system and affiliated caregivers. Systems, through their agreements and contracts, clarify the following: • the hours that providers are available for care; • the number of children caregivers are allowed to accept; • the rates providers are paid; and • whether providers may serve children who have not been referred by the system. This last point is especially important. When systems restrict the enrollment of homes to children referred to the caregiver by the system, the system is said to have an exclusive use agreement. Twelve of the 22 programs studied in the NDCHS had such agreements, which yield a number of benefits to the system. First, they allow the program to control the number of children in care at any time. Compliance with group composition requirements is thus assured. Se- cond, as caregivers often depend upon system staff to help with problems in such areas as child discipline, conflicts with parents, and fee payments, pro- grams with exclusive use agreements may limit potential difficulties of the care situation to their own enrolled children and families. Third, services per- 110 formed by the system, such as food reimbursement and field trips, are more easily managed when only system children are served. In operation, the ex- clusive use agreement also binds the caregiver more closely to the system in that it makes her entirely dependent on the system for her income. As a conse- quence, programs with exclusive use agreements appear to manage more closely-knit groups of providers. One disadvantage of exclusive use programs is that in some states such systems have very little flexibility in responding to the needs of families. For example, systems that serve only Title XX-eligible children may be forced to terminate a child if the family's income increases over the eligibility ceiling, making them ineligible for subsidized care. Such a situation sometimes leads parents to refuse a raise because the increased salary will make them ineligible for subsidized day care and thus effectively decrease net family income. An equally important consequence of this type of situation is greater instability of care for children who must move to a new arrangement if their family is dis- qualified for care through the day care system. When providers join systems, whether they contract their services or are paid as employees, and whether or not they have exclusive use agreements, they are expected to accept three types of responsibilities: to provide a safe and adequate caregiving environment; to develop and maintain caregiving skills; and to perform recordkeeping tasks. The selection of responsible and suitable providers is a major issue for new as well as established systems, since turnover and system growth continually create a demand for new recruits. A considerable staff investment is made in these new caregivers to insure that they can and will provide quality care. In a sense, system staff wear two hats. On the one hand, the local day care regulating agency frequently transfers much of the responsibility for regula- tion and monitoring of the home to the day care system.* On the other hand, system staff act to support their caregivers both professionally and emotion- ally. Systems screen providers for personal characteristics and physical health. Directors reportedly prefer applicants who are flexible, warm, loving, enjoy children, and have physical stamina. Motivations for applying are always probed. Directors are not particularly concerned with the potential caregiver's age, education, or experience, although they prefer experience in raising children. Another important criterion is that the provider be able to complete the required paperwork. Systems inspect and approve all prospective homes, whether or not the pro- vider is already licensed. System staff check on such features as cleanliness, sufficient space for children to play, adequate exits in case of fire, and other specific safety aspects of the home. If a caregiver is unlicensed, most systems can approve the home to care for children while it is affiliated with that *In fact, some systems themselves certify providers to care for children in lieu of state licensing. In this case, if the caregiver leaves the system she is no longer legally entitled to care for children in her home. Ill system. Because directors realize that convenience in transporting children is important to parents, they select homes which are convenient for their clients: close to the child's home; close to the parent's place of business; or somewhere in between. Equally important to systems in ensuring that caregivers provide quality care is the training that the system provides. Although all systems claimed to train providers, the range of topics, frequency of sessions, requirements for at- tendance, and importance placed on training varied considerably. On average, providers were offered five hours of training per month. Although providers were expected to attend training, one-third of the programs did not require at- tendance. When attendance was required, the attendance rate was approx- imately 85 percent. Where programs did not require attendance, only 50 per- cent of the providers regularly attended. Topics discussed at sessions are varied. They cover such areas as nutrition, community resources, child development, recordkeeping, health and safety, parent participation, art, activities for children, family day care as a business, insurance and taxes, problem-solving, role playing, observations of child care in centers, and development of providers' self-esteem. Since providers and parents occasionally have disagreements over child-rearing practices and lifestyles, topics related to cultural sensitivity are sometimes integrated into the training. Family day care systems perform a number of other useful services for pro- viders in addition to training. They distribute supplies, loan out safety equip- ment, and pay for liability insurance. They provide substitutes for caregivers who are ill and occasionally assign helpers. They are responsible for billing the government for reimbursement and may collect parent fees where appropriate. The provider may then be paid by the system rather than by the government funding agency (which frequently delays payments) or the parent (who occa- sionally does not pay). When the system pays providers for child care, they are usually paid in a regular and consistent fashion. Child Care Food Program To participate in the Child Care Food Program, a family day care home must be affiliated with a sponsoring organization. The term "sponsor" when used in relation to the Child Care Food Program does not necessarily denote the family day care system as we have discussed it elsewhere in this report. CCFP sponsorship can be an arrangement whereby an agency submits an ap- plication and reimbursement claims for a group of homes but performs none of the other services family day care sponsors typically provide, such as place- ment of children, collection of child care fees, provision of fringe benefits or training in child development skills. At the time of the study and continuing to the present, there have been relatively few family day care home systems and, consequently, a miniscule percentage of family day care homes are currently eligible to participate in the Child Care Food Program. In only a few states (notable three of those in our 112 study, California, Pennsylvania and Massachusetts) has there been a demonstrated government interest in developing systems as a vehicle through which to deliver publicly funded care. However, the Child Care Food Program has increasingly been the motivating factor in the creation of systems and in the decision of many providers to become affiliated with systems where they do operate. In several states the social service, human resources or public welfare agency of the state has taken on the role of sponsor, acting as a con- duit for CCFP funds because no "real" sponsors exist. The new CCFP regula- tions have taken note of the de facto exclusion of homes from program par- ticipation because of the lack of sponsorship and have authorized start-up funds to new or existing family day care systems to enable them to recruit up to 50 new or additional homes. At the time of our data collection 5 of the 22 systems in our study— 4 of them in California — did not participate in CCFP. This was a function of the newness of many of the systems but also reflected the reluctance of many pro- gram directors to take on what was viewed as the onerous recordkeeping responsibilities of the CCFP as it was then constituted. Since that time, for a variety of reasons, participation has been substantially increased. For exam- ple, today all California systems receiving public monies are required to par- ticipate in the CCFP, and participation in other states in virtually universal. In 1980, three years after the initial NDCHS system interviews, extensive discus- sions with state licensing offices, child care advocacy groups, welfare depart- ments and participating systems in states which were known to contain systems identified very few systems which did not participate in the CCFP. This in- creased participation on the part of the family day care systems does not reflect a change in CCFP as much as a growing knowledge of and familiarity with the CCFP in the child care world. When asked about the benefits of family day care system membership, pro- viders frequently mentioned the food program and the fact that it helps them to provide nutritious, high-quality food to children in care. Although the food program was originally intended to increase the nutritional intake of children from low-income families, all children in a day care home must surely benefit. To the extent that the provider must think through and plan her meals with the nutritional guidelines of USDA in mind and to the extent that the additional money is used to upgrade food, then meals for all children are improved. Services to Children and Families Most systems place children by setting priorities among parent needs, child needs and provider preferences. In most instances, systems are concerned about the match between parents and providers and encourage them to meet beforehand. In general, a home must have available space (children are not shifted from one home to another to make space), and must be conveniently located. Another major service of sponsoring agencies is the administration of child care subsidies through Title XX, allowing approximately 90 percent of their 113 children to receive free or reduced-fee child care. Occasionally, systems are capable of extending this service to those who do not qualify for federal assistance through scholarships and reduced fees supported by private and other public sources. This major administrative service opens the door for many families to receive the range of social services which frequently ac- company subsidized child care. (As indicated above, exclusive use systems often have less flexibility in this regard.) Only a few of the 22 systems directly administered some form of health or dental services to children or providers. These few had nurses or specialists who performed visual and dental screening, immunization, hearing and developmental tests. Five programs offered no help in arranging or providing medical screening or delivery. Of all systems, only three took on the full cost of providing such services. In an effort to encourage parent involvement, seven systems organized parents' advisory committees, but of the seven only three had active parent groups. Of these three, two served single teenage parents who are considered a high risk group. The remaining system served primarily private-fee parents, unlike most in sponsored care. Overall, then, parent involvement is minimal. Often parents work days and are unwilling or unable to become involved in the management or operation of the system. Most directors expressed a desire to encourage more participation, but were not certain that parents would par- ticipate. Only one program transported children to and from day care homes as a regular service. Transporting young children (infants, toddlers and preschoolers) is not only time-consuming but may also, depending on the age of the children, require the assistance of an adult to handle each child. In most instances, when homes are located within the parent's neighborhood, transportation is left to the parent. Sponsoring agencies in general seek to provide children with the advantages of care in a home environment — individual attention, flexibility and a home- like atmosphere— while still offering many advantages of center care, such as trained caregivers, an environment conducive to the development of social and cognitive skills, and an array of support services and child care resources. Linda Carney, a 30-year-old White woman, is married and the mother of two. She is also a professional family day care provider who provides care to five children from five different families in her home every day. Mrs. Carney is affiliated with a sponsoring agency whose policy is to place children in family day care for one year before moving them into a center. All the children in this home are preschoolers three to five years old. This is the age group Mrs. Carney prefers to work with, finding them the easiest to care for. "It's not babysitting," she emphasizes, "it's teaching." All but one of the children in her care are from single-parent families. One of them is from a Russian-speaking family. Mrs. Carney feels that one of the toughest things about her job is the emotional problems that the children bring 114 with them. There are hardly any intact families, and a lot of the children are really disturbed by the turmoil in their family lives. "I've had seven kids in three years with really heavy problems that had to be put in that program [an early intervention program in the affiliated center]," she says. For this reason, she tries to provide the children with a lot of structure. Considering her training as a caregiver (given by her sponsor), her 54-hour work week and the paperwork demanded by her sponsor (which she does on her own time), Linda Carney feels that providers like herself are underpaid. She earns $29 per week per child. Her husband earns about $24,000 at his job in sales, so only about one-quarter of the family's income is from family day care. Mrs. Carney's work day begins at 7:30 a.m., when the first child arrives, and ends when the last one leaves at about 6:00 p.m. Early in the morning and later in the afternoon, the children watch TV, but the rest of the time Mrs. Carney supervises them closely or plays along with them, indoors and out. Only during their naptime does she get a chance to get away a little bit. Operating Budgets of Family Day Care Systems What does the operating budget of a family day care system look like? What are its sources of income and what are its expenses? How does the cost of pro- viding care to children compare with government reimbursements for that care? And finally, how do providers' earnings fit into these equations? Family day care systems receive revenue from a variety of sources, including federal and state governments, payments from parents, local matching funds, the federal Child Care Food Program and interagency donations of labor and supplies. Agencies receive the majority (69%) of their income from federal (Title XX) and state funding sources.* The next largest source of funding was in-kind contributions (13%). An additional 10 percent of income was derived from cash contributions and other funding sources, such as community block grants. Parent fees represented about 5 percent of systems' income and the Child Care Food Program accounted for another 3 percent. The value and sources of cash and in-kind contributions varied across systems. As programs become large and more established, they generate addi- tional contributions, either in cash or in-kind. Cash contributions come from such sources as United Way, Catholic Social Services, community block grants, fund-raising efforts and matching funds from city and state agencies. In-kind contributions represent either intra-agency sharing of staff, supplies, and space or are from sources outside the umbrella agency. Annual total resource costs** ranged from $80,832 to $913,647 with a me- dian of $180,457 across all systems. Labor costs represent the greatest expense of family day care systems; 88 percent of annual program costs consisted of administrative personnel expenses and provider payments, and only 12 percent involved nonlabor expenses. *In California, the State Department of Education provides child care subsidies for family day care under Chapter 344, Statutes of 1976, Assembly Bill 3059. **The term resource costs refers to cash costs plus the value of in-kind contributions. 115 To compute the hourly cost of child care, annual cash costs were divided by annual child-hours. Costs per child-hour ranged from $.72 to $1.56 with a me- dian of $1.18. When in-kind contributions are added to cash costs, resource costs per child-hour range from $.79 to $2.46 per child-hour with a median of $1.21 . Although contributed resources added considerably to the value of care within specific programs, overall the median costs of care excluding such con- tributions would only drop from $1.21 to $ 1 . 1 8 per child-hour. Government reimbursement rates ranged from $0.59 to $1.53 per child per hour, with a median of $0.95. Most systems receive between $0.90 and $1.20 daily per child. The method of establishing reimbursement rates between systems and funding agencies is unclear. Although directors interviewed stated that rates were determined by a number of factors such as prior years' rates, projected annual budgets, cost of living increases (i.e., inflation), local price indices, planned service delivery, expansion and the agency's reputation as a child care delivery system, no programs were quantifiably evaluated by fund- ing sources to determine per-child rates. As noted above, only one of the 22 programs paid providers actual wages, offering them full employee status. To obtain estimated daily earnings per pro- vider, average provider rates per system were adjusted for the average number of children per home in each system. Based on this calculation, providers earned an average of $20.14 per day for child care, with a range from $9.69 to $37.18. This range reflects the combined variability across systems in base rates and group size policies. Few caregivers earn more than $23 daily and few earn less than $10 daily. A second way to view provider earnings is to consider the total income earned annually. Using the systems' total expenses for provider wages, annual earnings calculated for sponsored providers averaged $4573 for child care. Earnings ranged from $1640 to $7817. In general, system providers tended to earn more than their independent counterparts and more than providers of unlicensed care. Nevertheless, the vast majority of system providers earned wages considerably below the Poverty Line for 1977 ($6000). Indeed, only the 25 percent of providers with earnings above $5900 have incomes above the Poverty Line, and almost one-third do not earn minimum wages. None earned an income as high as the Department of Labor Low Income Budget (10,000 in FY 1977). Providers' hourly rates are established in a variety of ways. Directors spoke of establishing rates using rates of other child care systems, or those of in- dependent caregivers as a reference. Caregiver rates are also influenced by pro- posal budgets drawn up to negotiate government reimbursement rates. Therefore, systems which are reimbursed at higher rates from government sources pay providers higher rates for caregiving. Functional Cost Analysis As previously discussed, the unit costs of care and cost components vary by program. A closer look at these costs and their components is made possible 116 through a functional cost analysis of seven tasks performed by family day care systems. Figure 8 presents a breakdown of hourly costs by function. As the figure shows, $.65 (52%) of the mean $1.25 resource cost is spent on direct caregiving. Another $.28 (or 22%) is applied to administrative and overhead costs. Food program costs represent $.08; an additional $.08 was spent for licensing and monitoring; $.05 for training programs; $.04 for provider and child intake; and $.08 for social services and transportation programs. Figure 8 Functional Breakdown of Hourly Resource Costs 22% Administration & Overhead 52% Direct Caregiving 4% Training Program 3% Provider & Child Intake 6% Social Services 7% Food Program To get a clearer picture of the cost implications of services provided by spon- sors, core costs of program administration — direct caregiving, administration, overhead and food service — were isolated from supplemental service costs (see Figure 9), providing a measure of resources needed to operate a "no frills" package of sponsored family day care. Supplemental service costs include pro- vider training, transportation, social services and regulatory functions such as home approval and monitoring. Core costs account for 81 percent of all costs, with an hourly cost of $1.01. Supplemental costs constitute the remaining 19 percent, and have an hourly cost of $.24. Supplemental Services and In-Kind Contributions Two important findings emerged from our analyses of supplemental ser- vices. First, supplemental service costs were relatively small compared with the 117 Figure 9 Core Versus Supplemental Resource Costs Core tasks, 81%: direct caregiving, administration, overhead, and food program Supplemental tasks, 19%: licensing, monitoring provider training, social services and transportation essential core program costs. Therefore, core services clearly emerge as the most critical cost component to consider in understanding differences in cost allocation across programs. At the same time, supplemental services are not insignificant from a cost perspective, particularly if mandated on a broad scale. The relationship between supplemental services and in-kind contributions has an important implication for future federal funding for all child care pro- grams which are dependent upon in-kind contributions to meet critical needs and mandated requirements. Federal regulations require that family day care homes maintain information regarding special health precautions for children such as diet, medication and immunizations. Similarly, homes must provide information to parents concerning social services available in the community; systems must ensure that homes meet these regulations or must assume the responsibilities themselves. Currently, noncash resources represent significant and critical resources in meeting federal standards for supplemental services. Thus, in one respect, leverage of such resources represents a cost savings to the government. However, donated resources are not necessarily predictable and stable and thus cannot be projected to remain uniformly and universally available at cur- rent levels in the future. Rather, programs tend to compete locally for such limited free or third-party-paid resources, and the nature of the competition for these resources changes with changing federal, state and local priorities. 118 When considering cash costs and total resource costs, the policymaker should not necessarily favor the expansion of a lower-cost program that receives high levels of donations or meets regulations for services through donations, if maintenance at such levels of donations cannot be assured. Similarly, in the rate-setting process, proposed provisions of mandated ser- vices and proposed levels of government funding should take into account pro- jected total resource costs and the availability of such resources over the con- tract period. Direct Child Care Costs Versus Administrative Costs An important measure of the efficiency of systems in providing core services is the ratio of direct child care costs to administrative costs. This analysis com- pares the proportion of resources used to deliver direct child care services (such as wages for providers) to the proportion of total resources needed for general administrative support (such as non-caregiving staff salaries, space and materials needed to manage the program). Figure 10 displays resource costs per child-hour for direct care, for sup- plemental services and for administration. The first bar in the figure represents the average cost by functions across all systems. To generate the remaining two bars, the sample was divided into two groups: those with low administrative costs per provider and those with high administrative costs per provider. These two bars represent the average functional costs within each of the two subgroups. Figure 10 Average Functional Costs: Administration, Direct Care and Supplemental Services $1.60 $1.40 - $1.20 $1.00 - $ .80 $ .60 $ .40 $ .20 .0 Admin. $.28 Direct Care $.73 Supp. $.24 Cash Costs $1.13 Donations $.12 Admin. $.18 Direct Care $.72 Supp. $.27 Cash Costs $1.08 Donations $.09 Admin. $.40 Direct Care $.74 Supp. $.20 Cash Costs $1.19 Donations $.15 All systems (total resource cost = $1.25) Systems with low admin, costs per provider (total resource cost = $1.17) Systems with high admin, costs per provider (total resource cost = $1.34) 119 The average resource cost per child-hour across all systems is $1.25. Of this amount, $.73 or 58 percent is allocated for direct care costs, while $.24 (36%) is allocated for supplemental services. Administrative overhead costs at $.28 require 22 percent of total resources. As indicated, only $.12 of the total of $1.25 is generated from in-kind resources. These resources are devoted to sup- plementary services and cover roughly half the costs for supplemental services. In comparing the two graphs on the right-hand side of Figure 10, it is clear that there is little difference between the two groups in terms of dollars allocated per child for direct care and for supplemental services. Given the dif- ference in total resource costs ($1.34 versus $1.17), programs with lower ad- ministrative costs are delivering roughly the same child care and services. The difference between the two groups is largely attributable to substantially larger administrative costs per child incurred in the high-cost group. On the average, systems with higher administrative dollars per provider are spending twice as much to administer systems than those in the lower category. This finding sug- gests that higher reimbursement rates and greater costs are associated with higher staff costs and more administrative staff, but with no appreciable in- crease in services or the child care hours provided. To verify the relationship between administrative costs on the one hand and direct care and supplemental services on the other, the two subsamples were compared on a number of program dimensions. Overall there are no signifi- cant differences in service delivery, although the high administrative cost sam- ple provides more family services than the other group. The large difference in per-caregiver administrative costs is not outweighed by more direct care or supplemental services. Providers tend to be paid higher rates per child in the high-cost group, but comparative annual earnings for providers fall within the average range for all systems' providers. Furthermore, there are no significant differences in staff wages and benefits across the two groups, although directors in systems with high administrative costs tend to be paid somewhat more. Based on earlier findings, it can be hypothesized that they also have more experience and use this to generate more in-kind resources. Summary Because the sample of family day care systems reported on here is small, the reader is cautioned that our findings cannot be generalized to settings not covered by the study. For example, very large systems are not included. Never- theless, this descriptive study has provided an opportunity to learn what elements comprise a family day care system, to examine the key relationships between costs and services, to explore the nature of resources, and to describe allocation of resources to common system functions. This chapter has provided a case illustration of the trade-offs that can be made, given fixed resources, between administration and direct care for children, as well as between core functions and supplementary services. These data raise questions for the future funding of systems by investigating the rela- 120 tionship between funding and the establishment of rates. Most important, the discussion of costs and rates cannot be divorced from the issues of what caregivers can and are willing to earn as family day care providers. Without work performance standards and quantified measures for services, the rate-setting process for family day care systems will remain idiosyncratic. In areas of the country where reimbursement rates are set at the low end of the spectrum, provider rates will remain below $.75 per child-hour. Total costs will approximate reimbursement rates unless other sources of income are iden- tified to augment government subsidies and unless directors capture a share of the available community in-kind resources. Clearly there is no systematic method within or across states for setting reimbursement rates beyond the cyclic renegotiations between family day care systems and the state. Systematic methods for setting rates, however, cannot be achieved unless agreement is reached in each state regarding which services— and how much service— the state is willing to purchase. Without such standards, rates will continue to fluc- tuate based on what individual programs claim they are doing, assessed against individual program budget estimates, with no benchmarks with respect to allowable administrative costs or minimum provider wage rates. 121 Chapter 10 CONCLUSIONS AND RECOMMENDATIONS Family day care providers play an essential role in rearing the nation's children. The image of the two-parent family, wife at home with the children, remains in the forefront of our national consciousness. But the reality is quite different. Women have, in increasing numbers, entered the work force, mak- ing day care an ever more pressing issue for them and their families. More and more, the daily life of children is in the hands of family day care mothers. However, our persisting vision of the stereotypical family has long focused the nation's attention and research on the traditional family. Consequently, little has been known about the role of the day care provider in children's lives, despite the many millions of children who are each day cared for in family day care homes. It was with an understanding of the importance of family day care to American children and their families that the National Day Care Home Study was undertaken. A principal mandate of this research was to provide objective information needed to intelligently respond to exigencies of the changing American family. The present volume is a brief introduction to the richness of the patterns that have been uncovered. Along with the accompanying detailed research volumes in this series, it is intended to provide information necessary to the formulation of our national policy toward children and families. Although few could fail to be impressed with how well the "natural" family day care system currently functions, findings of the National Day Care Home Study lead to several conclusions about ways in which family day care could be supported to the benefit of children, parents and caregivers. The following recommendations are presented in this light. They could variously be im- plemented at the federal, state or local level to facilitate and improve the func- tioning of our child care system. Recommendations Related to the Supply and Demand of Family Day Care The demand for day care is expected to continue increasing over the next two decades, despite the declining birth rate, because increases in women's labor force participation will more than offset this decline. The rate of increase is expected to be particularly high for women with children under three. 123 Until recently, substitute care for children under three has been supplied primarily by members of the extended family, either in the child's own home or in the home of a relative. With the diminishing availability of relative care, due, for example, to increased population mobility, other sources of day care must be utilized. Since parents have shown a distinct preference for family day care over centers or other group facilities for infants and young toddlers, it is expected that the non-relative family day care market will absorb most of the increased demand for care of children under three. Recommendation 1: Promote the Growth of Family Day Care Supply to Meet the Increased Day Care Demand A. Demand for Infant and Toddler Care. The steadily growing labor force participation of women with children under three is dramatically increasing the number of infants and toddlers who need full-day care. Parents report that in the absence of suitable in-home care the most appropriate day care setting for children of this age is family day care. However, because the demand for infant and toddler day care on a relatively massive scale is a recent phenomenon, there has been little organized effort to increase the supply of care for these children. Consequently, parents in all income brackets currently find it difficult to find day care for children under three. To meet this demand we recommend encouraging growth in the supply of family day care for infants and toddlers. The public needs to be informed that, by and large, family day care has been found to provide a stable, warm and stimulating day care environment which caters successfully to the developmen- tally appropriate needs of the children in care; that parents who use family day care report it satisfactorily meets their child care needs; and that the cost of this care is reasonable. Second, additional public funds should be made available to support needed family day care slots for income eligible families with young children. At present, relatively limited public resources are allocated to family day care, while day care centers, which receive the bulk of day care dollars, predominantly serve children aged 3-5. There is thus little public money allocated to the care of infants and toddlers. Consequently, if in- come eligible families with young children are to become self-supporting, resources will be needed for the care of their young children, preferably in family day care, the care of their choice. Furthermore, since the 1980 HHS regulations* limit enrollments in homes caring for children under two years of age to levels below those allowed for older children, the care of young children is often not economically feasible (as costs are high and potential income is low). In order to encourage caregivers providing subsidized care for infants and toddlers, government reimbursement rates should be set at a higher level for care of children under two years of age. B. Demand for School-Age Care. When in-home care is not available, the most popular form of care for school-aged children is family day care. ♦At the time this report was published, the 1980 HHS regulations had been delayed by Congress until July 2, 1982. 124 Relatively few such children are found in day care centers, school-based facilities or other formal settings. Strikingly, however, the National Day Care Home Study has found that most family day care for these children is provided in informal, unregulated settings and that proportionally few school-aged children are found in regulated or sponsored day care. This unequal distribu- tion of school-age care probably reflects unmet demand for regulated and sponsored care for children in this age group. It occurs in part as a by-product of the former federal day care regulations which counted school-aged children in the same way as younger children in computing enrollment limits, even though the schoolers were only in care part-time. Consequently, many regulated providers did not find it economically practical to care for these children. The 1980 HHS family day care requirements have substantially altered this economic imbalance by allowing for an additional school-aged child in most homes. Supply of care for these children could be further increas- ed by raising the government reimbursement rates for school-aged children. Recommendation 2: Promote the Development of Day Care Systems We recommend that the development and expansion of family day care systems be emphasized as one of the principal means of providing subsidized day care in a family day care setting. Under the rubric of systems, we mean to include the wide variety of public and private agencies currently sponsoring family day care homes — private social service agencies, religious organizations and agencies of state and local governments. The trend toward organization of family day care into systems is a fairly re- cent development but one with important implications for day care programs and policies. Although public and private day care systems together account for a minute proportion of all family day care homes, they already account for a substantial percentage of all family day care subsidized through Title XX of the Social Security Act. In addition, systems act as conduits for all food sub- sidies provided through the USDA Child Care Food Program; and it is often the case that state day care resources are also channeled into family day care through family day care systems. With the growth of these systems, family day care homes are for the first time gaining access to the wide range of day care resources (such as caregiver training, health screening and other social ser- vices) which have in the past been largely confined to day care centers. The National Day Care Home Study has also found that day care systems play an important role in promoting quality care by maintaining desirable enrollment levels, monitoring regulatory compliance, training caregivers, pro- viding technical assistance to the caregiver and providing a vehicle for parent involvement. The organization of family day care homes under the umbrella of family day care systems thus has advantages for each of the participants— the government agency that subsidizes or regulates the care, the caregiver, the parent, the child and the community as a whole. The agency benefits because the system becomes administratively responsible for regulating the caregiver, monitoring 125 the quality of care, providing supplementary services and handling the com- plex issues of reimbrusement. The caregiver benefits because the system refers children, provides her with assistance in handling administrative respon- sibilities, pays her on a regular basis and provides her with the training and technical assistance needed to improve her caregiving skills. The parent benefits by having the agency help in finding an appropriate home for the child, provide substitute caregivers in case of illness, and arrange for a variety of supplemental services to the family. The child benefits from appropriate placement, from improved nutrition in programs subsidized under the Child Care Food Program, and from the skilled care of a trained caregiver. Finally, an increased emphasis on the development of family day care systems will help reduce the isolation of family day care from the rest of the day care commun- ity, thereby increasing its overall visability and accessibility, particularly for families needing subsidized care. Recommendation 3: Improve Community-Based Support for Parents and Caregivers A striking feature of family day care is its isolation and lack of visibility in the community. Parents report difficulty in locating family day care and pro- viders are often unable to replace children who leave the day care home. This is a disadvantage both to parents looking for care and to the caregivers, who fre- quently would like to care for more children than are currently enrolled. We therefore recommend increasing community-based family day care sup- port structures such as day care information and referral centers. Such centers serve to disseminate needed day care information throughout the community and by doing so help both parent and caregiver. The parent is assisted both in identifying available homes and in making an informed selection once the home is found. The caregiver obtains greater access to parents in search of child care. This helps her maintain her enrollment levels and thus assures a steady flow of income from caregiving. Further, information and referral centers provide the caregiver with practical advice about running a family day care home, advice typically available only to caregivers affiliated with family day care systems. The opportunities to obtain such positive benefits might well induce many currently unregulated providers to become part of the visible family day care network. Recommendations Related to Quality in Family Day Care Despite the widespread commitment to quality in family day care, it is not always clear what quality means or how to achieve it. Family day care is characterized by many cultural, community and family patterns. Part of the great richness of this type of care is its responsiveness to the needs of in- dividual families and children, its informal family-like style, its ability to ac- commodate children of different ages and various group configurations. 126 Just as there is no one definition of quality for family day care which will suffice, so there appears to be no single route to its achievement. Sometimes extensive experience as a parent and caregiver results in high individual skill. Study findings also indicate that effective caregiving skills can be learned through training, and monitoring can help insure quality from the standpoint of determining that standards are met. The following set of recommendations acknowledges the need for a multifaceted approach, taking each of the several known contributors to quality into account. Recommendation 4: Continue to Regulate Group Size and Age Mix to Protect Young Children, Limit Caregiver Burden and Create Flexibility for School- Aged Children The National Day Home Study findings provide substantial support for cur- rent federal policy related to group size and age mix: • Data on group composition established that the number of children and their age distribution in the family day care home do make a dif- ference in the kinds of experiences and opportunities children have in family day care. • The NDCHS showed that parents support government regulation of group composition. • A limit of approximately 6 children (depending on age mix) in family day care homes appears to be reasonable. Fully 90 percent of all day care homes studied in the NDCHS cared for 6 or fewer children — regardless of regulatory status, ethnicity, length of time in business, amount of education, training, experience, or any other fac- tor. Fifty percent had three or fewer children. • Regulations which allow fewer infants and toddlers but which also pro- vide for additional school age children are appropriate and consistent with actual practice. • Infants tended to be found in smaller homes, either alone or with one other child. Larger homes generally included school-aged children who are in care for only a few hours a day. Futhermore, many providers who cared for more than 6 children spaced them throughout the day so that large numbers were not in the home at any one time. • Caregivers preferred slightly larger group sizes than they actually had. However, longitudinal data from Los Angeles documented that group composition remains stable over time and that caregivers usually replace their children on a one-for-one basis to maintain the same type of group even as individual children change. These findings support the notion that family day care providers structure their homes to be manageable and comfortable for them. We therefore support the emphasis on group composition as expressed in the 1980 HHS Day Care Requirements. We feel that it is appropriate to set limits on group size and that these limits should take into account the ages of children in care. In addition, we underscore once again the importance of a 127 public information campaign to inform parents and caregivers of the implica- tions for the children of various enrollment patterns. Recommendation 5: Increase the Availability of Caregiver Training The National Day Care Home Study has shown that training does make a difference in the kinds of experiences and opportunities available to children in family day care homes. Caregivers themselves expressed a desire for training. This was true across the board, regardless of regulatory status or cultural iden- tification. Parents also prefer that their children's caregivers be trained. This preference was the second most cited qualifications factor for parents— only experience as a parent was more important. In reality, however, very few fam- ily day care providers have been trained and those who have are most likely to be in sponsored settings. We therefore recommend that sound caregiver training programs be iden- tified or developed to address the wide range of skills and knowledge needed by family day care providers. With the understanding that family day care serves children from birth through school age and that it is closely tied to the community and the cultural values of the families it serves, we recommend that training programs be composed of separate modules specifically designed to address a variety of typical family day care arrangements. In this way, com- plete training programs of differing orientation, scope, and specific content can be developed by combining appropriate modules. To facilitate the development and implementation of efficient and effective systems for the delivery of training to local providers, the Federal government should also provide direct hands-on technical assistance to states and local communities. Particular emphasis should be placed on public awareness of caregiving training, its benefits and its availability. Finally, for those providers who are not required to and do not choose to participate in formal training, there should be an ongoing media campaign and wide public distribution of information about child development and child care with concrete, helpful suggestions for operating a family day care home. Recommendation 6: Establish a Family Day Care Credentialling System Because family day care is such a diverse and fragmented field with few pro- fessional affiliations, it is difficult for the public, including both consumers and providers, to assess caregiver skill — one of the most fundamental ingre- dients of family day care quality. It is not enough to know that the provider has the proper number of children or that physical dangers are under control. What adults do with children is fundamental and supremely important to their well being. A credentialling system to certify family day care providers who meet objective criteria will help emphasize caregiver skill, provide a standard of excellence for parents, lend credibility and a sense of professionalism to family day care and help establish a badly needed career ladder for family day care providers. 128