NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) and Institute of Medicine (US) Committee on Integrating the Science of Early Childhood Development; Shonkoff JP, Phillips DA, editors. From Neurons to Neighborhoods: The Science of Early Childhood Development. Washington (DC): National Academies Press (US); 2000.

Cover of From Neurons to Neighborhoods

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

Show details

11Growing Up in Child Care

Second only to the immediate family, child care is the context in which early development unfolds, starting in infancy and continuing through school entry for the vast majority of young children in the United States. It is the setting in which most children first learn to interact with other children on a regular basis, establish bonds with adults other than their parents, receive or fail to receive important inputs for early learning and language development, and experience their initial encounter with a school-like environment. Early and extensive enrollment in child care has become the norm in U.S. society. Indeed, if children were only sporadically or briefly exposed to child care, it would not be the visible policy issue that it is today.

In 1994, 10.3 million children under the age of 5 were in child care while their mothers worked, including 1.7 million infants under 1 year of age (U.S. Bureau of the Census, 1997). The vast majority of 5-year-olds are in kindergarten (88.5 percent in 1995) (Hofferth et al., 1998). Younger children have also been enrolling in center-based child care, preschool, and pre-kindergarten programs at increasing rates so that, by 1997, 45 percent of 3- and 4-year-olds and 22 percent of children younger than 3 were in these types of programs (Capizzano et al., 2000; Ehrle et al., 2000). But enrollment in child care begins long before this. In 1999, the National Household Education Survey, which asks all families about nonparental child care arrangements regardless of the employment status of the mother, reported that 61 percent of children under age 4 were in regularly scheduled child care, including 44 percent of infants under 1 year, 53 percent of 1-year-olds, and 57 percent of 2-year-olds (see Figure 11-1).

FIGURE 11-1. Percent distribution of newborn to 4-year-old children in nonparental care on a regular basis, by age, 1999.


Percent distribution of newborn to 4-year-old children in nonparental care on a regular basis, by age, 1999. SOURCE: Unpublished tabulations from the 1999 National Household Education Survey; generated for the committee by DeeAnn Brimhall, National Center (more...)

This is a dramatic change from the recent past. With it have come both growing acceptance of child care as supplementing rather than competing with parental care and persistent worries about the effects of child care on children's development. The dilemmas that today's parents are facing are not new, however. Decisions about the care and supervision of young children are among the oldest problems faced by human society (Lamb, 1999; Rossi, 1977). Over the history of family life and across cultures, mothers have had multiple duties that have necessitated sharing the handson care of their infants and toddlers with others, primarily other women relatives and older children (Lancaster and Lancaster, 1987; LeVine et al., 1994; Weisner and Gallimore, 1977).

What is new is the rapid growth in reliance on paid care by nonrelatives in center-based settings and the expansion in public subsidies for child care. While parents and relatives continue to provide vast amounts of early child care, rapid growth in reliance on center-based arrangements as the primary source of child care has occurred for children of all ages, accompanied by a decline in the use of home-based care by nonrelatives. The ramifications of welfare reform—the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996—for child care are also changing the landscape in unprecedented ways. Prior to the 1996 legislation, states were prohibited from requiring recipients who were single parents caring for infants to participate in work-related activities. As of June 2000, 14 states have used the new flexibility granted by the legislation not to exempt automatically from work requirements parents whose youngest child is less than 1 year old (and most of them require work when the infant reaches 3 months of age). An additional 23 states require mothers receiving benefits to work when their children reach age 1 (State Policy Documentation Project, 2000). Moreover, for single mothers, over half of the states require 30 or more hours of work per week. As a result, the population of children in child care is likely to include more very low-income infants than has ever before been the case.


What do we mean by child care? It is not just day care, given the growing numbers of children who require supervision while their parents work nontraditional and shifting hours. It is also not just care. Beneficial outcomes for children in child care are associated with settings that provide both nurturance and support for early learning and language development. Accordingly, previous distinctions between “early education” or “pre-school” and “day care” have unraveled. In fact, child care may be seen as providing a number of services, including the provision of nurturance and learning opportunities for children, preparation for school, support for working parents and reduction of poverty, respite care in child welfare cases, and access to supplemental services such as vision and hearing screening, developmental testing, feeding programs, and even parent support and literacy programs (Fein and Clarke-Stewart, 1973; Lamb, 1998; Scarr and Eisenberg, 1993).

While many of these purposes are complementary, the distinction between child care as a developmental program for children and child care as a support service for working parents continues to guide different emphases in policy debates (Blau, 2000). This is most apparent with respect to the differing attention given to issues of the quality of care supported by different policies. For example, 25 percent of all new funds for Head Start, which emphasizes developmental goals, is set aside for quality improvement initiatives. In contrast, only 4 percent of the funds for the Child Care and Development Fund (CCDF)—the major source of child care support tied to welfare reform—is dedicated to quality improvements. There are indications, however, that the political divide between these two tiers of child care policy making is becoming less distinct, as funding streams for state prekindergarten, Head Start, and CCDF-funded child care programs are increasingly being merged at the federal, state, and program levels (Kagan and Cohen, 1996; Raden, 1999; Schulman et al., 1999).

The research reviewed in this section covers the broad array of programs and services that provide for the care and early development of young children while their parents work or, for other reasons, rely on others to provide care for their children on a regular basis. Recognizing that substantial controversy surrounds nomenclature in this area of research, practice, and policy, we use the term “child care” throughout this report to encompass the blend of care, nurturance, and early education that the best child care provides. We focus on naturalistic studies of community-based child care settings—ranging from grandparent care to preschool programs—given that these, in all of their diversity, constitute most of the child care for children in the United States. They also are the focus of concern regarding the developmental effects of child care. Within this literature, we emphasize studies that have examined the effects of child care net of family influences on development. We also include evidence from planned interventions, discussed more extensively in Chapter 13, when they supplement and sharpen knowledge about child care. Research on school-age child care is not included in this synthesis, given the focus of this report on children prior to school entry (for excellent recent reviews on school-age child care, see Vandell and Posner, 1999; Vandell and Shumow, 1999). Following a brief discussion about the timing of entry into child care and factors that impinge on this decision, we synthesize the literature on the effects of child care on both the mother-child relationship and child development. We then discuss research on the ingredients of quality care that promote beneficial development, the availability and distribution of higher-quality arrangements, and child care for children with disabilities.


Parental decisions about child care are an important component of parental influence in the early childhood years. The first decisions about child care that face new parents are whether and when to place their child in nonparental child care and what specific arrangement to select. Corresponding to the rapid growth in labor force participation of mothers with children age 1 and younger (see Chapter 10), the majority of parents now enroll their children in child care during the first year of life. National survey data reveal that, as of the mid-1990s, approximately 1.7 million infants under 1 year of age were in child care while their mothers worked (Hofferth et al., 1998; U.S. Bureau of the Census, 1997). Data from the NICHD Study of Early Child Care (see Box 11-1 for a description of the study), which is the only prospective study of parents' child care decisions, further reveal that enrollment in child care occurs very early in the first year. In this study, 72 percent of the infants experienced some nonparental child care in the first year of life, with an average age at entry of 3.31 months (NICHD Early Child Care Research Network, 1997b). About three-quarters of those who entered care during the first year of life entered prior to age 4 months and they were in care for an average 28 hours per week. The picture these data provide is thus one of very early entry into extensive child care.

Box Icon

BOX 11-1

The NICHD Study of Early Child Care and Youth Development. Aware of the growing use of child care and the increasing public and policy concern about this issue, the National Institute of Child Health and Human Development (NICHD) of the U.S. Department (more...)

The extent to which the high use of child care at early ages reflects parents' desire to return to work quickly or financial constraints on their ability to remain at home with their infants remains an open question. Pertinent information is available, however, regarding access to and use of family leave benefits, as well as about families who adjust their work schedules to curtail their reliance on nonparental child care for their babies.

The Role of Parental Leave

It is well documented that use of infant care is substantially lower in countries that have generous parental leave policies (Kamerman and Kahn, 1995). Prior to passage of the Family and Medical Leave Act (FMLA) in 1993, the United States was the only industrialized country without a federal law guaranteeing a job-protected maternity leave. In the absence of federal legislation, 23 states had passed leave laws that cover both private-and public-sector workers, but with varying provisions (Commission on Family and Medical Leave, 1996). The federal law requires employers with 50 or more workers to offer a job-protected family or medical leave of up to 12 weeks to qualifying employees (those who have worked at least 1,250 hours in the previous year) who need to be absent from work for reasons that meet the terms of the law, including the need to care for a newborn or a newly adopted or new foster child.

It is estimated that these provisions of the FMLA leave 89 percent of all private-sector work sites and 53.5 percent of the nation's private-sector employees uncovered (Commission on Family and Medical Leave, 1996). Nevertheless, the law appears to have had a major impact on the number of companies who are now offering job-protected leaves for maternity and other family and medical reasons, as well as on increased use of leave by employees (Waldfogel, 1999a, 1999b). Much of this increased use has been among men who appear to be using the leave for “other” family and medical reasons (i.e., for reasons of their own health or to care for an ill family member). There is also evidence that more leave is being used by women with infants as a result of the FMLA, although this appears to be due not so much to more women taking infant care leave as to women taking more leave (Klerman and Leibowitz, 1998; Rossi, 1998).

The law does not require the leave to be paid, but it does require that employers who provide health insurance coverage to continue to do so during the leave period. This raises questions about who avails themselves of leave and who does not. National survey data collected by the U.S. Department of Labor following implementation of the FMLA (see Cantor et al., 1995) reveals that only 17 percent of covered employees took leave during 1994-1995 and an additional 3.4 percent indicated that they needed but did not take leave. Two-thirds of workers who needed but did not take a leave indicated that they could not afford the associated loss of wages. Parents who have access to parental leave benefits and can afford to make use of them do so, suggesting that the enrollment of very young infants in child care is not entirely voluntary. Results from the NICHD Study of Early Child Care mentioned earlier indicating that the families who placed their infants in child care at the youngest ages (before 3 months) were heavily or entirely dependent on the mother's wages to escape poverty, and that many had previously been poor or dependent on public assistance, lend support to this possibility (NICHD Early Child Care Research Network, 1997c).

Parents' Arrangements for Child Care

The arrangements that parents make for the care of their children span every conceivable combination of care by mothers, fathers, and others, the complexity of which tends to get lost in efforts to categorize and portray them. In some countries, the predominant form of child care is sibling care (Harkness and Super, 1992; Nsamenang, 1992; Zeitlin, 1996). In Cameroon, for example, infants and toddlers are usually cared for by preadolescent girls, often older siblings or relatives, as part of the girls' preparation for their adult roles. After weaning, the peer group becomes the ubiquitous socializer and caretaker of children. While sibling care is much less common in the United States, it does occur.

Most young children in the United States are, however, with adults. Figure 11-2 provides information on the care arrangements used by families where the primary caretaker of the child was employed in 1997 (Capizzano et al., 2000; Ehrle et al., 2000). There are two very different ways of looking at these data. One view focuses on the large extent to which infant and toddler care, and to a lesser extent preschool care, remains within the family, shared equally by parents and other relatives. The other view focuses on the extent to which parents rely on nonfamilial care and move their children rapidly into formal group care arrangements.

FIGURE 11-2. Current distribution of care for infants and toddlers, and preschoolers with employed mothers, 1997.


Current distribution of care for infants and toddlers, and preschoolers with employed mothers, 1997. SOURCE: Capizzano et al. (2000).

As has historically been the case, a surprisingly large number of employed parents with young children do not rely on others for child care at all. In 1997, for example, a little over one-quarter of families with at least one employed parent and an infant or toddler under age 3 relied primarily on parental child care while the primary caretaker was working. Hispanic families are somewhat more likely than others to rely on parents for infant and toddler care (32 percent did so in 1997; Ehrle et al., 2000), but it is also very common among white (27 percent) and black families (22 percent).

Child care provided by fathers (while mothers work), for example, has crept upward from 15 to 21 percent of all infant and toddler care arrangements between 1977 and 1994 (U.S. Bureau of the Census, 1997). Fathers provided one in four of the first child care arrangements made for the infants in the NICHD Study of Early Child Care (NICHD Early Child Care Research Network, 1997b). While reliance on parent care is much more common among two-parent families in which only one parent works or both parents work part-time (44 percent of families with children under 3 and 30 percent of families with children ages 3 to 4), it is also surprisingly common among two-parent families in which both parents work full-time (16 percent of families with children under 3 and 12 percent of families with children ages 3 to 4) and in one-parent families that get by with parttime employment (26 percent of families with children under 3 and 7 percent of families with children ages 3 to 4) (see Figure 11-3). Clearly, a considerable number of parents are making the effort to care for their own children, usually at home, perhaps at considerable cost to their family incomes.

FIGURE 11-3. Reliance on parent care by family structure and extent of employment, 1997.


Reliance on parent care by family structure and extent of employment, 1997. SOURCE: Unpublished tabulations from the 1997 National Survey of America's Families; generated for the committee by Gina Adams and Jennifer Ehrle, The Urban Institute.

Once parents turn to others for assistance with child care, grandparents and other relatives are the caregivers for many families, including 27 percent of children under age 3 and 17 percent of 3- and 4-year-olds. Hispanic families are particularly likely to rely on relatives for infant and toddler care (39 percent of Hispanic families do so; Capizzano et al., 2000; Ehrle et al., 2000), compared with black (27 percent) and white families (25 percent).

At the same time, as noted above, there has been extremely rapid growth in reliance on center-based care not only for preschoolers, but also for infants and toddlers (see Figure 11-4). The share of children under age 3 in child care centers, preschools, Head Start programs, and other early childhood education programs tripled between 1977 and 1994, from 8 percent to 24 percent of children with employed mothers (U.S. Bureau of the Census, 1982, 1997). In contrast to patterns of family-based care, center-based care is used much more by black and white families than by Hispanic families, with the largest discrepancies appearing for infant and toddler arrangements (rates of use are 30, 24, and 10 percent, respectively; Capizzano et al., 2000; Ehrle et al., 2000). The increased use of center care has been accompanied by declining use of family child care providers. Nevertheless, as of 1997, 39 percent of infants and toddlers and 59 percent of preschoolers were in center-based or family child care arrangements with nonrelatives (see Figure 11-2), revealing the rapid movement of children into formal care settings and peer groupings during the earliest years of life.

FIGURE 11-4. Growth in use of center-based care, 1977-1994.


Growth in use of center-based care, 1977-1994. SOURCE: U.S. Bureau of the Census (1982, 1987, 1997).

In sum, vast numbers of infants spend substantial portions of their time in child care, often starting within a few months after birth. While much of this very early care remains within the family—with parents who are juggling their work schedules and with relatives—young children move rapidly into nonrelative care as they enter the toddler and preschool years. Although we know virtually nothing about the factors that impinge on parents' decisions about when to first rely on child care, it appears that these decisions are affected by a complex mix of factors including access to parental leave, the capacity to forgo wages for a period of time, new policies requiring work from mothers formerly dependent on public assistance, and the availability of child care arrangements (including sharing care between two parents) with which the parents are comfortable. The emerging evidence on these issues belies the hesitancy and ambivalence that accompanies new parent's decisions about infant child care and renders questions about the consequences of child care for young children especially compelling. In this context, issues concerning equity of access to family leave benefits become important, as do questions about the extent to which families in differing circumstances (e.g., those without a partner available to share child care responsibilities) feel that they are able to exert their preferences regarding when and how they arrange for the care of their infants.


Two concerns have guided research on the developmental effects of child care. The first focuses on the mother-infant relationship and asks, “Will this relationship be harmed or diminished in significance as a result of the daily separations that are entailed when a baby is placed in child care?” This concern is not unfounded. Child care, insofar as it reduces the amount of time available for the mother to learn the baby's signals and rhythms, might also adversely affect her ability to respond sensitively to the baby and establish a secure attachment relationship (see Brazelton, 1986). The other concern focuses directly on the children: “Will the young child's cognitive, language, and social-emotional development be compromised as a result of spending time in child care?” Today, this concern is riveted on infants and toddlers, for whom early and extensive enrollment in nonfamilial child care is a relatively recent phenomenon. The National Research Council summarized the evidence on these issues a decade ago (National Research Council, 1990). The intervening decade of research has both confirmed and expanded on the earlier panel's conclusion that the effects of child care derive not from its use or nonuse but from the quality of the experiences it provides to young children. (For additional, recent reviews of research on child care see Lamb, 1998; Love et al., 1996; Scarr and Eisenberg, 1993, and Smith, 1998.)

Child Care and the Mother-Infant Relationship

Evidence from child care research of the 1990s is reassuring to those who have been concerned that child care might disrupt the mother-infant relationship. Not only does the mother remain the primary object of attachment for infants in child care (Ainslie and Anderson, 1984; Farran and Ramey, 1977; Howes and Hamilton, 1992; Kagan et al., 1978), but also the attachment relationship appears to be largely protected from possible negative effects emanating from early entry into and extensive hours of care, as well as poor-quality care (NICHD Early Child Care Research Network, 1997a; Roggman et al., 1994; Symons, 1998). The primary influence on the attachment relationship derives not from child care but from the sensitivity of the care that is provided by the mother (namely, her supportive presence, positive regard, and lack of intrusiveness and hostility). This is equally true for children experiencing very little child care and children experiencing a lot of child care (NICHD Early Child Care Research Network, 1998b).

But the mother-child relationship is not necessarily unaffected by child care. In addition to studying the attachment between mother and infant, researchers have made direct observations of mother-infant interaction. Although many studies find no effects of child care on mother-infant interaction, some report positive effects and still others report that child care appears to create or compound problems that are seen in these interactions. Looking again at the NICHD study, infants and toddlers in more hours of child care, regardless of its quality, experienced somewhat less sensitive mothering and were less positively engaged with their mothers than other children who were not enrolled in child care (NICHD Early Child Care Research Network, 1999a). Among infants and toddlers enrolled in child care, however, those in higher-quality arrangements (regardless of hours in care) experienced greater maternal sensitivity. Moreover, the negative relation between the amount of child care and maternal sensitivity and child engagement with mother was not of sufficient magnitude to disrupt the formation of a secure infant attachment. When considered in conjunction with the data presented in the previous chapter, suggesting detrimental effects of maternal employment in the first year of life, there would seem to be cause for concern about early infant care, particularly in light of its highly variable quality in the United States (discussed below).

A number of other studies have found that when very young children (i.e., infants and toddlers under 2 years of age) are exposed to risk factors at home and to extensive or poor-quality early child care, their odds of experiencing insensitive mothering increase (Belsky et al., 1996c; Clark et al., 1997; Tresch Owen and Cox, 1988). The direction of effects underlying these findings is not yet clear: that is, there have been some suggestions that early reliance on child care undermines the mother's ability to respond sensitively to her child and, as a result, diminishes the child's involvement with her mother (see Clark et al., 1997; NICHD Early Child Care Research Network, 1999a; Stifter et al., 1993; Tresch Owen and Cox, 1988), but other studies fail to find these associations between early child care and maternal sensitivity (Burchinal et al., 1992; Egeland and Hiester, 1995; Goldberg and Easterbrooks, 1988; Gottfried et al., 1988; Rabinovich et al., 1986; Stith and Davis, 1984; Zaslow et al., 1985). It may also be the case that less sensitive mothers are more likely to enroll their infants in child care at a very early age, although this hypothesis has not been tested.

Child care can also protect children from family-based risk. This has been a primary rationale for early intervention programs that provide highquality center-based child care for children living in poverty and for children in the child welfare system. Naturalistic studies of typical child care have also demonstrated protective influences. For example, mothers participating in the NICHD study who were living in or near poverty and whose infants were in full-time, high-quality child care were observed to show more positive involvement with their 6-month-olds (i.e., spontaneously vocalizing, responding verbally to the child, voicing positive feelings, hugging, kissing, praising) compared with similarly poor mothers who were rearing their babies at home or were using full-time, lower-quality infant care (NICHD Early Child Care Research Network, 1997d). Others have found that child care can protect infants and older children from the detrimental effects of both poverty (Caughy et al., 1994) and maternal depression (Cohn et al., 1986, 1991).

In sum, despite persistent concern about the effects of child care on the mother-infant relationship, the weight of the evidence is reassuring, with the possible exception of emerging findings regarding very early, extensive exposure to care of dubious quality. Mothers themselves play the lead role in determining the quality of their relationship with their infants, toddlers, and preschoolers. If anything, the child care research of the past decade has enhanced appreciation of the potent influence of parents on early development. When child care effects are examined net of parental effects on child outcomes, parent's behaviors and beliefs show substantially larger associations with their children's development than do any features of the child care arrangement. These efforts to control for family influences when examining how child care affects child development might appear to be an obvious and straightforward approach. Nevertheless, it has only become common in the past decade. In fact, even with extensive controls for family influences on development, it is impossible to be assured that we are capturing the effects of child care untainted by influences that result from the fact that families with different features (e.g., higher incomes) are able to place their children in child care with different features (e.g., higher-quality care).1

When child care is found to be associated with the mother-child relationship, the link is as likely to be positive as it is to be negative. The challenge now facing those who study child care is to clarify when child care protects children from family-based risk (such as poverty, maternal depression, high levels of conflict), when it compounds risk, and when it poses risks to children who otherwise are growing up in supportive home environments. In other words, we need to ask more complicated questions regarding how child care intersects with what transpires at home (Hoffman, 1989; Zaslow et al., 1985).

Effects of Child Care on Children's Development

Ultimately, questions about child care turn on its consequences for child development. Under what conditions does child care contribute to or undermine children's social skills, emotional well-being, and readiness for school? The answer is “it depends,” but a great deal more is known about what it depends on than was known a decade ago.

One of the most consistent and ubiquitous findings in this literature links the quality of child care that children receive to virtually every measure of development that has been examined. While hours of care, stability of care, and type of care are sometimes associated with developmental outcomes, it is the quality of care and, in particular, the quality of the daily transactions between child care providers and the children for whom they are responsible, that carry the weight of the influence of child care on children's development. This conclusion, based largely on correlational studies of typical child care, is confirmed by experimental evidence linking enrollment in very high-quality early intervention programs to both short-and longer-term outcomes in both academic attainment and prevention of delinquency for high-risk children (see Barnett, 1995; Currie, 2000; Shonkoff and Meisels, 2000; Yoshikawa, 1994, 1995; as well as the discussion in Chapter 13). What remains to be understood is whether investments in quality that fall substantially short of the levels entailed in the intervention programs can produce meaningful benefits not just for high-risk children, but for all children. Let's look at the evidence.

Effects on Cognition and Language

As a result of concerns about school readiness (or, in the case of early intervention programs, hopes for promoting readiness), emerging competencies in cognitive and language domains have been a long-standing focus of study in child care research. Outcomes that have been assessed range from IQ and general developmental levels to specific learning and communication skills.

The strongest and most compelling evidence regarding the developmental effects of high-quality child care on early cognition and language has come from experimental studies of planned early interventions for economically disadvantaged children or for those at risk of developmental problems. The findings from this literature are consistent. Intensive, high-quality, center-based interventions that provide learning experiences directly to the young child have a positive effect on early learning, cognitive and language development, and school achievement (Barnett, 1995; Brooks-Gunn et al., 1994; Burchinal et al., 1997; Feagans et al., 1995; Lamb, 1998; Ramey and Ramey, 1998; Roberts et al., 1989). Sometimes these effects dissipate during the early school years, but the impacts of some programs have been found to continue well into the school years and even into adulthood (Campbell and Ramey, 1994; Currie and Thomas, 1995; Lazar and Darlington, 1982; Luster and McAdoo, 1996; McLoyd, 1997; Yoshikawa, 1994, 1995). Effect sizes in this literature range up to 1.0 standard deviation for outcomes for preschoolers.2 The early intervention literature further indicates that the strongest effects of high-quality care are found for children from families with the fewest resources and under the greatest stress.

High-quality care in the infant and toddler years is also associated with children's cognitive and linguistic development in the correlational research on typical child care settings (Burchinal et al., 1996; Galinsky et al., 1994; Howes and Rubenstein, 1985; McCartney, 1984; Peisner-Feinberg and Burchinal, 1997; Peisner-Feinberg et al., 2000). One of the few studies that provided effect sizes reported that they ranged from .09 to .14 for associations between child care quality and cognitive and language outcomes for 3-year-olds (NICHD Early Child Care Research Network, 1999c). The evidence associating quality of care and early cognitive and language outcomes is striking in its consistency. The results are often but not always stronger for children from lower-income families and those whose mothers have relatively low levels of education (Peisner-Feinberg and Burchinal, 1997). Sometimes improved language and learning outcomes are short-lived (Chin-Quee and Scarr, 1994; Deater-Deckard et al., 1996) and sometimes they endure into the school years (Andersson, 1989; Broberg et al., 1997; Burchinal et al., 1995; Field, 1991; Larsen and Robinson, 1989; NICHD Early Child Care Research Network, 2000; Peisner-Feinberg et al., 2000; Rosenthal, 1994; Vandell and Ramanan, 1992). Of central importance to cognitive and language outcomes is the verbal environment of the child care setting (McCartney, 1984; NICHD Early Child Care Research Network, 2000). As with mother care, child care providers who are both supportive and provide more verbal stimulation have children in their care who show advanced cognitive and language development.

In light of the experimental evidence on center-based early intervention programs, it is interesting that evidence is emerging from nonexperimental studies of more typical child care suggesting that cumulative experience in high-quality, center-based care starting in the second year of life may be particularly beneficial for cognitive development (Broberg et al., 1997; Hartmann, 1995; NICHD Early Child Care Research Network, 2000). Some studies find that center-based care is especially beneficial for children from low-income families (Caughy et al., 1994), but others find that all children benefit regardless of their family background (NICHD Early Child Care Research Network, 2000). What might children be getting in child care centers that they are not getting in other settings? One of the features that distinguishes higher-quality from lower-quality care with regard to early cognition and language is the amount of language stimulation that child care teachers provide (McCartney, 1984; Melhuish et al., 1992; NICHD Early Child Care Research Network, 2000). Center-based teachers, who are more likely to have received specialized training in early development and more education generally than providers in other child care settings (NICHD Early Child Care Research Network, 1996, in press b). As a result, they may talk more with children and respond to their efforts to communicate in precisely the ways that foster early language and cognitive skills, but this speculation requires empirical study.

Effects on Social and Emotional Development

Efforts to understand how child care affects children's social-emotional development have assessed a vast array of outcomes that tap children's self-regulatory behavior, their cooperation with and attachments to adults, their social skill (or lack of it) with other children, and the developmental level of their social interactions. For virtually every outcome that has been assessed, quality of care shows positive associations with early social and emotional development (see NICHD Early Child Care Research Network, 1998c and reviews by Lamb, 1998; National Research Council, 1990; Scarr and Eisenberg, 1993) after family influences on development are controlled, albeit to varying degrees. The experimental literature on early intervention also has demonstrated significant effects on young children's social skills and, in particular, on reduced conduct problems (Yoshikawa, 1994, 1995). Indeed, it is in the realm of preventing delinquency in adolescence and early adulthood that the strongest economic effects of early intervention appear to be focused (see Chapter 13). When children enter high-quality child care earlier and spend more time in these arrangements, positive effects on social competence can continue on into the elementary years (Peisner-Feinberg et al., 2000) and even preadolescence (Andersson, 1989; Field, 1991), although this is not consistently the case.

The child's relationship with his or her child care provider seems to play an especially important role with regard to social-emotional development. Children form secure attachments to their child care providers when they are stable and these attachments, in turn, are associated with adaptive social development, just as they are for children and parents (Howes et al., 1992; Oppenheim et al., 1988; Peisner-Feinberg et al., 2000; Pianta and Nimetz, 1991; Sroufe et al., 1983). Howes and her colleagues have found, for example, that children who are securely attached to their providers show more competent interactions with adults and more advanced peer play (Howes and Hamilton, 1993; Howes et al., 1988, 1994), both during the child care years and on into second grade (Howes, 2000).

Others have found associations between the stability of child care providers in center-based programs and the quality of children's interactions with their providers (Barnas and Cummings, 1994), as well as their social competence with peers, active engagement with materials in the classroom, and vocabulary levels (Howes et al., 1992). As reviewed in Chapter 7, the stability of the peer group may matter as well. Children who remain longer with the same group of children are more peer-oriented and less solitary over time than those whose peer groups have changed frequently (Galluzzo et al., 1990; Harper and Huie, 1985; Holmberg, 1980; Howes, 1988a, 1988b) and they are friendlier toward peers in distress (Farver and Branstetter, 1994).

In sum, the positive relation between child care quality and virtually every facet of children's development that has been studied is one of the most consistent findings in developmental science. While child care of poor quality is associated with poorer developmental outcomes, high-quality care is associated with outcomes that all parents want to see in their children, ranging from cooperation with adults to the ability to initiate and sustain positive exchanges with peers, to early competence in math and reading. This conclusion derives from experimental research on high-quality interventions for children at risk, as well as from the weaker correlational designs that assess a broader range of quality and a broader distribution of children. The stability of child care providers appears to be particularly important for young children's social development, an association that is attributable to the attachments that are established between young children and more stable providers. For cognitive and language outcomes, the verbal environment that child care providers create appears to be a very important feature of care.

The influence of child care is not as large as the influence of the family environment, but it emerges repeatedly in study after study, using different measures, and for children of different ages and living in different circumstances. Most studies of typical child care have not, however, followed children on into elementary school, let alone into adolescence. This is an important missing piece in the child care literature that is needed to understand the conditions in schools, families, peer groups, and communities that sustain positive, early child care effects. The studies of typical child care also remain open to criticism, discussed earlier, based on the difficulties associated with the fact that parents select their children's child care settings. To address this criticism, research on typical child care settings using experimental and other stronger designs is needed. In particular, a firmer understanding is needed of the causal impacts of differing amounts and types of investment in child care quality that, for reasons of political feasibility, fall short of providing high-quality interventions for all children.


Volumes of both scholarly and popular material have been written about the ingredients of high-quality child care (Lamb, 1998; Love et al., 1996; Phillips and Howes, 1987). Although some have argued that the factors that parents care about differ from the features of child care that researchers tend to study, in fact, the differences appear to be more a matter of terminology than of substance (Hofferth et al., 1998). For example, parents want their children to receive lots of individual attention and to be exposed to materials and interactions that will prepare them for school. Researchers in search of the central features of quality care have identified the relationship between the child and the care provider and the amount of cognitive and language stimulation provided over the course of the day as especially critical. In general, three tiers of variables have been examined in studies of child care quality: the child-provider relationship, the structural features of care, and the surrounding community and policy context. They can be viewed as nested levels in which the quality of child-caregiver interaction is affected by the quality that characterizes the structural features and community context of care.

The Child and the Caregiver

Quality of care ultimately boils down to the quality of the relationship between the child care provider or teacher and the child. A beautiful space and an elaborate curriculum—like a beautiful home—can be impressive, but without skilled and stable child care providers, they will not promote positive development.

Young children whose caregivers provide ample verbal and cognitive stimulation, who are sensitive and responsive, and who give them generous amounts of attention and support are more advanced in all realms of development compared with children who fail to receive these important inputs (see Lamb, 1998; Smith, 1998). This conclusion applies to infants, toddlers, and preschoolers and also applies to all forms of child care, ranging from relatives to center-based programs (NICHD Early Child Care Research Network, 1998c, 2000). Stability and skill appear to go together. More stable providers have been found to engage in more appropriate, attentive, and engaged interactions with the children in their care (Raikes, 1993; Rubenstein et al., 1977; Whitebook et al., 1990). It is not a coincidence that the high-quality intervention programs that have generated strong experimental evidence of positive developmental effects have employed highly qualified staff and experienced virtually no teacher turnover (National Research Council and Institute of Medicine, 2000).

Stable child care providers are rare, however. Turnover rates among them (including those who change settings as well as those who leave the field) are among the highest of any profession that is tracked by the U.S. Department of Labor (U.S. Bureau of Labor Statistics, 1998), hovering at 30 percent per year. By comparison, 6.6 percent of public school teachers and 21 percent of home health aides leave their jobs each year. Multisite, observational studies of child care centers have reported turnover rates in the 1990s ranging from over 40 percent (Whitebook et al., 1990, 1997) to 25 percent (Phillips et al., 1994). In 1977, the annual turnover rate among center-based providers in these same sites was 15 percent (Coelen et al., 1979). The authors of a multisite study of home-based providers (Kontos et al., 1995) reported that 30 percent of care arrangements provided by relatives were no longer available after a year, 25 percent of unregulated family day care providers had gone out of business, and 8 percent of regulated family day care providers were no longer operating.

Structural Features of Care

The next tier of quality consists of features that are associated with warm, sensitive, and stimulating interactions on the part of child care providers and teachers. Solid evidence has documented associations among the provider's behavior, her self-reported training and education, and the immediate context in which she works, including ratios, group size, and the adult work environment (Lamb, 1998; Love et al., 1996; Smith, 1998). Some intriguing recent evidence suggests that the staff-child ratio may be relatively more important for infants and toddlers and that the educational level of the provider may become more important as children move beyond the infant years into toddlerhood and beyond (NICHD Early Child Care Research Network, 1996, in press b).

Both formal education levels and recent, specialized training in child development have been found quite consistently to be associated with high-quality interactions and children's development in center-based, family day care and even in in-home sitter arrangements (Dunn, L., 1993; Fischer and Eheart, 1991; Kontos et al., 1994, 1995; Lamb, 1998; NICHD Early Child Care Research Network, 1996, in press b; Smith, 1998; Whitebook et al., 1990). Caregivers with more child-centered and less authoritarian beliefs about childrearing have also been found to provide warmer and more sensitive care (NICHD Early Child Care Research Network, 1996, in press b; Phillips et al., 1987a). Among home-based providers, the choice of child care as a career (sometimes called “intentionality”) has been associated with higher-quality care (Kontos et al., 1995). Experience as a child care provider, in contrast, shows a much less consistent relationship to quality care (Dunn, L., 1993; Galinsky et al., 1994; Kagan and Newton, 1989; Kontos, 1994; Kontos and Fiene, 1987; Ruopp et al., 1979; Whitebook et al., 1990).

The context within which caregivers work has been examined most often using measures of ratio and group size that capture the demands on an individual child care provider's time and capacity to provide sensitive care to her young charges. The ratio of children to caregiver has held up over time as one of the most sensitive indicators of quality care in all settings as, to a somewhat lesser extent, has group size (Burchinal et al., 1996; Galinsky et al., 1994; Lamb, 1998; NICHD Early Child Care Research Network, 1996, in press b; Phillipsen et al., 1997; Ruopp et al., 1979; Smith, 1998; Whitebook et al., 1990). Importantly, it appears that fairly minor changes in ratios and group sizes can affect the quality of care that young children receive. For example, 6-month-olds who are the only child in care have been found to receive significantly more positive care-giving than infants in settings in which one additional child is present, even when the additional child is the caregiver's own (NICHD Early Child Care Research Network, 1996). Infants in centers with ratios of three or fewer children per caregiver have been found to receive significantly more sensitive and appropriate caregiving (Howes et al., 1992), and to score one standard deviation above those in centers with larger ratios on a measure of communication skills, even after adjusting for family factors that affect development (Burchinal et al., 1996). The addition of two school-age children to family day care homes caring for infants, toddlers, and pre-schoolers was observed to result in less sensitive caregiving (Howes and Norris, 1997).

More recently, aspects of the adult work environment of child care, including provider wages and benefits, have been included in studies of child care quality. This research has revealed strong relationships, comparable to those found for training and ratios, between staff wages and child care quality in both center-based and family day care arrangements (Cost Quality and Outcomes Study Team, 1995; Helburn, 1995; Kontos et al., 1995; Phillips et al., 1991, in press; Scarr et al., 1994; Whitebook et al., 1997). Wages are also the primary, although not the only, determinant of staff turnover; when wages are increased, turnover declines (Whitebook and Bellm, 1999; Whitebook et al., 1997).

In light of this evidence, it is of concern that the average hourly wage of child care workers is $6.12 and that of family child care providers is $3.37 (U.S. Bureau of Labor Statistics, 1996). This is less than the hourly wage of parking lot attendants ($6.38) and bus drivers ($11.56), and substantially below the wages of kindergarten teachers ($19.16). Wages are not only low, but they have also not kept pace with inflation, and they often do not reflect the educational levels of child care providers. For example, in 1988, child care teachers in the National Child Care Staffing Study with some college education earned an average of $9,293 per year compared with the average wage of $19,369 for women with some college education in the civilian labor force (Whitebook et al., 1990).

The Community and Policy Environment

The final tier of quality consists of the broader community and policy environment in which child care operates. Important elements of this environment include the financing and regulatory structures that bear on the child care market, community-based planning systems, consumer education and involvement, systems for staff development and leadership training, and interconnections among providers working in different sectors of the market (Gormley et al., 1995; Kagan, 1993; Phillips, 1996). Child care regulations, which have been the focus of study in efforts to understand how the surrounding context of child care affects quality of care, appear to establish a floor of quality for regulated dimensions of care (i.e., ratios, group size), which, in turn, is associated with differing distributions of quality in states with more or less stringent regulatory provisions (Cost Quality and Outcomes Study Team, 1995a; Helburn, 1995; Howes et al., 1995b; Phillips et al., 1992). However, more stringent regulations may have the unintended effect of reducing the supply of regulated programs (Hofferth and Chaplin, 1998).

Voluntary systems may also be effective. Child care centers that voluntarily meet widely accepted guidelines for quality, such as those recommended by the American Public Health Association and the American Academy of Pediatrics (1992) provide better care, and the children in these programs show better outcomes than their peers in programs that do not meet these guidelines. For example, the mean school readiness scores for children in classrooms meeting none of the APHA/AAP standards was about 14 percentage points below the population norm; the scores for children in classrooms meeting all of the standards was just above the population average (NICHD Early Child Care Research Network, 1998c). Children in centers that met more of the standards had higher scores than did children in centers meeting fewer of the standards. In other words, there were no clear thresholds above which outcomes were markedly improved—more was better. Of course, we cannot ascribe the better outcomes directly to the standards. Centers meeting more standards may also be doing other things that foster development, and parent-driven selection bias may also be operating such that children who would do well in any case are more likely to be placed in high-quality settings. It is notable, however, that state child care standards fall far short of the APHA/AAP standards and vary enormously, from mandated ratios for infants ranging from 3 to 1 to 12 to 1 and for 3-year-olds ranging from 7 to 1 to 17 to 1 (for example Azure, 1996). Most states permit infants and toddlers to be cared for by staff who, on average, have not completed high school, have only had some general training in child development, and receive fewer than 5 hours of in-service training annually (Young et al., 1997).

In sum, quality is inherent in the child care provider, whether it is the grandmother, an unrelated sitter, or a center-based teacher. Critical to sustaining high-quality child care for young children are the providers' characteristics, notably their education, specialized training, and attitudes about their work and the children in their care, and the features of child care that enable them to excel in their work and remain in their jobs, notably small ratios, small groups, and adequate compensation. Regulatory and voluntary systems that support higher levels of quality on these dimensions are associated with variation in the quality of care that is found in given states, communities, and programs. Even small improvements in ratios and education are reflected in more sensitive, appropriate, and warm caregiving, suggesting useful targets for investments in quality. The success story provided by the U.S. Department of Defense's efforts to improve its child care programs attests to the feasibility of upgrading the quality of child care in the United States (see Box 11-2). It is important to recognize, however, that other dimensions of quality that are rarely measured (i.e., the leadership skill of the center director, the mental health and motivation of the caregiver, the stability of funding, characteristics of the families served) are, in all likelihood, important ingredients along with the structural dimensions of care that dominate the research literature (Blau, 1997, 2000). Without attention to some of these subtle, but potentially powerful, influences on quality, it is difficult to predict how much can ultimately be accomplished by policy actions that focus on only one or two structural dimensions of care.

Box Icon

BOX 11-2

Child Care for U.S. Military Families. The U.S. armed services oversee a child care system that serves more than 200,000 children every day at over 300 worldwide locations and includes families from all four branches of the military. The military child (more...)


Some child care settings offer children what they need to feel secure and loved, to learn, and to build social skills and friendships. Many do not. Virtually every systematic effort to characterize the quality of child care in the United States has found that about 10 to 20 percent of arrangements fall below thresholds of even adequate care (Cost Quality and Outcomes Study Team, 1995; Galinsky et al., 1994; Helburn, 1995; Whitebook et al., 1990). This is the case regardless of the type of care being examined. What do researchers see when they go into these settings? They see caregivers who more often ignore than respond to young children's bids for attention and affection, a dearth of age-appropriate or educational toys, and children who spend much of their time wandering aimlessly around, unengaged with adults, other children, or materials. Given the likely possibility that providers who offer extremely poor-quality care do not participate in research, these figures may actually be underestimates of the amount of poor-quality care that exists in this country.

In some cases, infants appear to get the poorest-quality care, but in other cases they have been found to get better care than older children, particularly when they are in a one-to-one arrangement with a competent caregiver. Even the NICHD Study of Early Child Care, which provides a more favorable portrait of child care quality than do other studies, reported that one in four infant caregivers were moderately insensitive, only 26 percent were moderately or highly stimulating of cognitive development, and 19 percent were moderately or highly detached (NICHD Early Child Care Research Network, 1996). Fewer than 20 percent of toddlers and preschoolers were in settings in which caregivers offered care that was “highly characteristic” of positive caregiving.

It is not unusual for basic safety to be compromised in the nation's child care settings, as illustrated by a 1998 Consumer Product Safety Commission (CPSC) study of 220 licensed child care settings. The study reported pervasive health and safety violations: two-thirds of the settings they visited had at least one safety hazard, including cribs with soft bedding, no safety gates on stairs, unsafe (or no) playground surfacing, and use of recalled products (Consumer Product Safety Commission, 1999). An earlier investigation conducted by the Office of the Inspector General (1994) found more than 1,000 violations in 169 child care facilities in five states. Among the hazards were fire code violations, toxic chemicals, playground hazards, and unsanitary conditions.

This range of quality becomes particularly worrisome when juxtaposed with evidence about who experiences better and worse child care in the United States. Children from poorer and more stressed homes receive lower-quality child care than other children (Howes and Olenick, 1986; NICHD Early Child Care Research Network, 1997c; Phillips et al., 1994). There is, however, one exception to this pattern. Among families using child care centers, the working poor and those whose incomes hover just above the poverty line receive poorer-quality care than either families living in poverty or families with solidly middle and upper incomes (NICHD Early Child Care Research Network, 1997c; Phillips et al., 1994). This counterintuitive finding appears to be attributable to differential access to child care subsidies and programs such as Head Start and other publicly subsidized arrangements that are available to the very poor, but not to families with somewhat higher incomes. Quality of care in these programs is significantly higher than in other community-based child care centers (Layzer et al., 1993; Phillips et al., 1994; Whitebook et al., 1990).

The link between subsidized care and quality care is not surprising in light of estimates of what it costs to provide high-quality child care. The cost of providing accredited3 center-based child care was estimated at $4,797 per child per year in 1988 ($6,764 in 1998 dollars) (U.S. General Accounting Office, 1990). A more recent analysis of the cost of care in Air Force child care centers, about 90 percent of which are accredited, estimated the per hour cost at $3.86 per child in 1997, which would amount to over $7,000 per year for 50 weeks of full time care (U.S. General Accounting Office, 1999). The average cost per child of Head Start was $5,021 in 1998—a largely part-day program serving 3- to 5-year-olds for 34 weeks a year.

Setting aside quality, the average cost (to families) of child care was $60.17 per week for children under age 5 and $66.39 per week for infants under 1 year in 1993. This amounts to costs of $3,609 for preschoolers and $3,982 for infants for full-year care in 1998 dollars (U.S. Bureau of the Census, 1995). The most thorough analysis of who pays the costs of providing center-based care (similar analyses are not available for other forms of care) found that parent fees cover less than half the full cost of care (Helburn, 1995). A sizable contribution toward the cost of child care (estimated at 20 percent of costs) consists of forgone earnings by child care providers who would receive substantially higher wages in other sectors of the labor market. A third of the costs (in 1993-1994) were paid by federal and state governments and other subsidies and contributions.

Even though many parents do not pay for child care, it represents a substantial financial burden to those who do pay4 and, in particular, to those who have meager incomes and lack subsidized care. This is not a small group. The vast majority of children with working mothers and family incomes below 200 percent of the poverty line receive no or almost no federal subsidies for their child care (U.S. Council of Economic Advisers, 1997). In 1998, only 15 percent of the children eligible for the Child Care and Development Fund—the major source of federal child care assistance for low-income families—actually received help through the program (U.S. Department of Health and Human Services, 1999d). Subsidies that lower the price of child care induce low-income mothers to work (Blau and Hagy, 1998; U.S. General Accounting Office, 1995) and lead to increased reliance on paid care rather than unpaid care, although not necessarily higher-quality care (Blau and Hagy, 1998; Hotz and Kilburn, 1992; Ribar, 1995).

Child care expenses are often the second or third largest item in a low-income working family's household budget. In 1993, for example, child care expenses averaged 18 percent of family income, or $215 per month, for poor families paying for care for a preschool-age child (U.S. Bureau of the Census, 1995). Average monthly costs for nonpoor families were higher in absolute terms—$329 per month—but lower as a percentage of the household budget—only 7 percent. The average share of income devoted to child care was even higher—at 25 percent—for families with incomes of less than $14,400. Thus, families with meager incomes not only spend substantially more of their income on child care, but also are priced out of higher-cost forms of care, namely centers and many licensed family day care homes, in many areas of the country (U.S. Department of Health and Human Services, 1999d). This is compounded for families with infants, for whom the cost of care is significantly higher (see above) compared with older children (U.S. Bureau of the Census, 1995).

While the type of care selected by a family is often a matter of personal choice, there is growing evidence that, without access to subsidies, low-income parents are often precluded from enrolling their children in more expensive center-based and other arrangements. Other factors come into play as well, including the high proportion of low-income mothers (41 percent; U.S. Bureau of the Census, 1997) who work nonday shifts and are largely precluded from using centers and regulated family day care homes (Hofferth, 1995; National Research Council and Institute of Medicine, 1995b; Siegel and Loman, 1991) and the low supply of center-based and other arrangements in low-income neighborhoods (Queralt and Witte, 1998). These constraints may be reflected in the results of a nationally representative survey of families using child care (Brayfield et al., 1995), in which 27 percent of parents with children under age 5 and incomes less than $15,000 expressed a desire to change their child care arrangements. Two-thirds reported a preference for care in centers, and 70 percent cited quality as the principal reason for wanting to switch.

In sum, the child care that is available to parents with young children in the United States is highly variable in quality, unlikely to offer stability, and supported primarily by parent fees. Several comprehensive studies have now reported that a sizable minority of children receive substandard care, and two federal investigations have found rampant safety and health violations in regulated programs. Indeed, the most characteristic feature of child care in the United States may not be what many have described as its typically mediocre quality, but rather the immense range in quality that is tolerated. The higher-quality programs are inequitably distributed and often beyond the reach of families with meager incomes, unless they are poor enough to receive heavily subsidized care and can adjust their work schedules to accommodate these arrangements. Finally, it is critical to recognize that prevailing fees for child care depend heavily on child care providers' low wages which often fail to reflect their educational attainments—a situation that fuels extremely high rates of turnover and instability for children and their parents.


Only a few decades ago, most children with disabilities were raised in foster or group homes or in specialized institutions. Today, nearly all children with disabilities are raised at home by their parents. As of 1996, a national health survey of households (1996 NHIS) found that 2.5 percent of children under 5 years of age, or 513,000 children, were limited in their activities and living at home. Half of these children experienced major limitations, such as mental retardation and cerebral palsy.5 Data from the U.S. Bureau of the Census reported by Brandon (submitted) indicate that nearly 4 percent of households included a preschooler with a disability.

This has turned attention toward the needs of working families with young children who have disabilities. As the 1996 welfare reform law affects a growing share of families in poverty, it is likely that even more mothers of young children with disabilities will be returning to work. It is well documented that children with a variety of special needs are overrepresented in poverty samples (Meyers et al., 2000). Consequently, the availability and quality of child care for children with disabilities is likely to become a more significant issue than is the case today. Unfortunately, there is very limited information about the child care arrangements for these children.

Like all families with young children, those whose children have a disability or special health care need are faced with the challenges of finding good-quality, affordable child care. But the inability or unwillingness of many child-care providers to accept children with disabilities (Berk and Berk, 1982; Chang and Teramoto, 1987), transportation and other logistical problems, difficulties with coordinating early intervention and child care services, and the scarcity of appropriately trained caregivers (Kelly and Booth, 1999; Klein and Sheehan, 1987) make the effort to find any child care a tremendous challenge for these families. One multisite study reported that 45 percent of mothers of an infant with a disability reported that they were not planning to work because they could not find child care, and 31 percent indicated that they could not find affordable child care (Booth and Kelly, 1998, 1999). The severity of the child's disability or illness greatly compounds these problems (Breslau et al., 1982; Warfield and Hauser-Cram, 1996).

Not surprisingly, the added caregiving demands of having a child with a disability lead to lower rates and fewer hours of employment among parents (overwhelmingly mothers) of these children compared with other parents (Brandon, submitted; Breslau et al., 1982; Jacobs and McDermott, 1989; Leonard et al., 1992; Wolfe and Hill, 1995). This is particularly true of families who have a severely disabled child or more than one child with a disability (Meyers et al., 2000), yet a large share of mothers of a moderately disabled child also report barriers to work. These relations hold even when other individual and structural factors that predict employment are taken into account. It also appears that mothers of children with disabilities are less likely to have reentered the labor force by the child's first birthday and are employed for fewer hours than mothers of typically developing children (Booth and Kelly, 1999).

When children with disabilities require child care, the expense to the family can be considerable. Recent survey and administrative data from California (Meyers et al., 2000) reveal that child care is the most common form of out-of-pocket expense for families with disabled children from birth to age 5 (with 25 percent of all families paying for child care), even more common than medical expenses. Child care is also the most expensive single category of expense, including medical expenses, for these families, with an average weekly cost of $141.87.

Surprisingly little is known about patterns of child care usage or the quality of care received by children with disabilities. Available evidence suggests that children with disabilities begin child care at older ages, are enrolled for fewer hours, are more likely to be cared for by relatives, including fathers, and less likely to be in child care centers than other children (Booth and Kelly, 1998; Brandon, submitted; Landis, 1992; Warfield and Hauser-Cram, 1996). One study reported that infants with disabilities received significantly poorer-quality care in child care centers than in child care homes or relative care, regardless of whether the centers provided early intervention services. Overall, however, approximately 60 percent of the infants were receiving relatively high-quality care. Moreover, the children in higher-quality care had more advanced motor development and higher adaptive behavior scores than children staying at home with their mothers at 30 months of age (Booth and Kelly, 1998, 1999; Kelly and Booth, 1999). Other studies have also reported benefits to children with disabilities that accrue from child care, as well as benefits to their families (Guralnick, 1976; Ispa, 1981).

In sum, despite the increasing influx of children with disabilities into child care, little is known about the conditions that support or hinder their access to care, their experiences in care, or how factors such as the type or severity of the child's disability or the child's family circumstances affect these issues. Even less is known about these issues from the perspective of child care providers, for whom anecdotal reports are beginning to reveal serious concerns with respect to the administration of medical procedures, inadequate training, and even explicit fears about children with disabilities. Much more research is needed on these concerns to inform parents, policy makers, and the wide range of practitioners who work with children with disabilities and their families.


The topic of care for young children cuts to the heart of conceptions of parental roles and responsibilities. Parents seeking a balance between providing economic resources for their families and providing care and nurturance for their children face competing pressures. Should they forgo income so a parent can remain home full-time with a young child? Should they arrange their jobs so they can combine work and child care without relying on others? Should they combine employment with nonparental child care? For some parents, these options represent real choices, but for others work is less a choice than an economic necessity, and for still others, work is now required. Nevertheless, a sizable minority of parents manage to care for their children during the earliest months and years of life without relying on others, despite the lost income that this may involve.

For the many parents who do arrange for nonparental child care, it is reassuring that child care is not the inevitable risk factor that some have portrayed it to be, nor does it replace parents as the major influence on early development. At its best, child care can be a significant source of nurturance, friendships, and early learning for the fortunate children in high-quality, stable arrangements. At its worst, however, child care can expose children to safety hazards, extremely unstimulating environments, and unresponsive supervision. Not surprisingly, the basic elements of high-quality care closely resemble the qualities of good parenting. Children's basic needs for consistent, sensitive, and stimulating care transcend the difference between home and child care. Moreover, when children's home environments fail to offer them this care, child care environments that do provide it can protect and promote their early development. By the same token, poor-quality child care can compound the consequences of problematic parenting. What remain to be specified for policy purposes are the dollar amounts and types of investments in quality improvements that are sufficient to produce meaningful improvements in developmental outcomes both for children living in high-risk situations and for children who are largely protected from these circumstances. This should be a high priority for future research on child care.

Safety hazards and settings that basically warehouse young children are inherently intolerable. But, even setting aside these programs, most of which refuse to participate in child care research, the wide range of care that is captured in research is associated with varying developmental outcomes. While the associations are seldom large, they are consistent and statistically significant, starting in infancy and continuing through the pre-school years, and, in some cases, on into the early elementary grades. When child care is of very high quality, as is the case for model early intervention programs, the positive effects can endure into the early adult years, particularly for children from the poorest home environments. However, the fortunate low-income children who have access to these programs are out-numbered by thousands of others who, for financial as well as other reasons, receive some of the poorest-quality care that exists in communities across the United States. Thus, many children who can benefit greatly from high-quality child care are unlikely to get it.

If young children were only sporadically or briefly exposed to child care, we might not need to be concerned about the portrait of child care quality and its associations with developmental outcomes that emerges from this review of research. But child care is an enduring fixture on the early childhood landscape, starting within the first few months of life, for substantial hours each day, and continuing up to school entry and beyond. Apart from the evidence that children's developmental trajectories are influenced by the child care they experience, the day-to-day quality of young children's lives is profoundly affected by the quality and continuity of their experiences in child care. It appears that even small improvements in ratios and training, and relatively modest compensation initiatives, can produce tangible improvements in the observed quality of care. But the larger need is for communities to create more viable systems of child care that do not tolerate unsafe and unstimulating settings, actively promote and reward high-quality care, stem the tide of staff turnover, and enable parents at all income levels to avail themselves of quality care for their children (Kagan and Cohen, 1996; National Association of State Boards of Education, 1991; National Research Council, 1990).



These selection biases, which arise from the fact that parents select their children's child care environments and do not do so randomly, can contribute to both over- and underestimates of associations between child care and child development (see Chapter 4, as well as Blau, 1999, and Duncan et al., 2000).


Effect sizes provide an estimate of the fraction of a standard deviation in the child outcome associated with a change in the given quality measure. Effects less than .3 are considered small, .3 to .5 as moderate, and above .5 as large (Cohen, 1988). There is, however, an emerging debate about how best to portray the magnitude of effects in this literature, with several experts now suggesting that effect sizes be interpreted not in absolute terms but in relation to other pertinent effects (McCartney and Rosenthal, 2000).


The National Association for the Education of Young Children administers an accreditation program for child care centers with well-specified criteria for “developmentally appropriate” care ranging from the structural features discussed above to required elements of teacher-child interaction to dimensions of the curriculum. Centers volunteer to participate, engage in an extensive self-study period, and are then visited by trained experts who assess the center's compliance with the accreditation criteria.


In 1999, 70.6 percent of parents paid for child care for their children age 4 years or younger. (These data are based on unpublished tabulations from the 1999 National Household Education Survey, which were generated for the committee by DeeAnn Brimhall, National Center for Education Statistics, U.S. Department of Education.)


The activity limitation data are based on unpublished tabulations from the 1996 National Health Interview Survey, which were generated for the committee by Paul Newacheck, University of California at San Francisco.

Copyright 2000 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK225555


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (5.0M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...