Child Care for Low-Income Families: Summary of Two Workshops (1995)

The Bureau of the Census (1995 ) introduced its most recent report on child care by stating that the preschool years are a time when children are most dependent on care providers' nurturance and supervisory skills. Numerous studies of child care quality have documented that the skills of the providers and the quality of the settings that are available to families in the United States are highly uneven. The quality of a sizable minority of arrangements is considered by many to be mediocre, at best, and is judged by some to be likely to compromise children's safety and development. Because good-quality and poor-quality care can be found in every type of arrangement, type of care should not be considered a proxy for quality of care. Variation in the quality of typical child care appears to affect numerous aspects of children's development assessed in the short term with available measures. The cumulative and long-term developmental and societal impacts of these effects remain to be evaluated.

Most of the research on child care quality and its effects on children has focused on families that are not poor. Research on preschoolers in low-income families has tended to examine the effects of early interventions, such as Head Start, rather than of more typical full-day community-and home-based child care.

The new wave of research presented at the workshops has attempted

to bridge child care and early intervention studies and, in many cases, to include families that span a broad income range. This more recent research has found that child care for low-income families encompasses a spectrum of quality comparable to that available to higher-income families, at least within center-based options. The literature has also begun to explore issues of equity of access across income groups to safe arrangements that offer learning opportunities and other experiences beneficial to children.

These studies also include informal care arrangements, previously neglected by researchers, among the types of care considered. They have captured a much wider age range than was previously the case, moving beyond studies of preschoolers to examine infant and toddler care, before-school and after-school care, and—significantly—to include a major, longitudinal study of child care that recruits families ranging from poor to wealthy at the time of their children's birth and follows them through the children's entry into school (the National Institute of Child Health and Human Development's Study of Early Child Care). Finally, the new research on quality typically involves multisite designs and much larger samples of children than was the case previously. Table 2 summarizes the basic features of the major, multisite studies of child care quality presented at the workshops.

This literature was examined at the workshops with three major issues in mind: (1) What is known about the quality of care that is available to low-income families across types of care, and how does it compare with the range of options available to higher-income families? (2) How is the development of low-income children affected by the range of quality to which they have access? (3) Given that low-income parents value aspects of quality when they search for child care, are they aware of and do they value what research indicates to be important dimensions of quality?

As with all families, low-income families face a range of child care options that vary from the unsafe and unstable to those that offer not only safe but also developmentally beneficial care and learning opportunities. A small, but significant, share of arrangements are consistently rated as poor-quality care. By the same token, most studies find very few settings at the high end of quality. Much of the data on quality of care presented at

the workshop encompasses families across all income ranges. This information is summarized first, followed by the evidence that was presented on quality for low-income families.

The newly released four-state study of Cost, Quality, and Child Outcomes in Child Care Centers (Cost, Quality, and Child Outcomes Study Team, 1995; hereafter referred to as the Cost and Quality Study) found that 12 percent of the over 200 classrooms that were observed offered less-than-minimal to inadequate levels of supervision, health and safety practices, nurturance, and learning materials and opportunities (ratings of less than 3 on a 7-point scale ranging from “inadequate ” to “excellent”). The younger the children served, the poorer the quality of care: 40 percent of infant and toddler rooms provided less than minimal levels of care. These classrooms are characterized by safety problems, poor sanitation practices, unresponsive (though rarely harsh or explicitly punitive) caregivers, and a dearth of toys and other child-related materials.

Margaret Burchinal's North Carolina study of infants from low-income and middle-income African American families in 23 center-based classrooms revealed that 14 of the 23 classrooms studied (over 60 percent) offered less-than-minimal levels of care (Burchinal et al., 1995).

The Study of Children in Family Child Care and Relative Care (Galinsky et al., 1994; hereafter referred to as the Family and Relative Care Study) observed 226 home-based child care providers across three states. The percentages of less-than-minimal care were 13 percent for regulated home-based providers, 50 percent for unregulated home-based providers, and 69 percent for relative care. All of the providers cared for children under age 6.

Data presented at the workshop from the NICHD Study of Early Child Care included the full spectrum of child care arrangements for 6-month-olds, ranging from care provided by children's fathers while their mothers work to care in licensed centers. Quality of care was assessed using multiple methods ranging from detailed observations of the adult interactions received by a specific 6-month-old child to setting-level quality ratings and provider interviews. (Given the range of arrangements assessed, the quality rating scale used in each of the studies described above was not appropriate for use in this study, making it difficult to compare findings across studies.) Across all settings, 19 percent of the infants received care from providers who were rated as moderately to highly detached and 26 percent received care from providers who were rated as moderately to

highly insensitive. These findings reflect observations of 576 infants across 10 research sites, 15 percent of whom were in father care, 17 percent in care by other relatives, 15 percent in in-home care by nonrelatives, 35 percent in family day care homes, and 18 percent in centers.

It is tempting, though premature, to attempt to integrate the findings from these recent studies of child care quality. The different studies included differing proportions of low-income children; for example, the Family and Relative Care Study disproportionately sampled arrangements serving low-income families, whereas the NICHD Study of Early Child Care did not. They also employed different data collection strategies and, in some important instances, different measures of quality.

What is evident across studies is that some young children are receiving quite poor-quality care based on observations by trained researchers of their care settings and of the behaviors of their providers. The percentages of settings about which concerns have been raised are seldom below 15 percent, suggesting that one in seven children receives poor quality care in available, typical child care arrangements. Further, the workshop participants cautioned that these data may actually underestimate the proportion of care settings that are of poor quality. In every study, a sizable number of providers refused to participate (in the Cost, Quality, and Child Outcomes study, for example, between 59 and 32 percent of the centers in the participating research sites refused to participate). Although there are many reasons why child care providers may not want to participate in research, one plausible reason is a concern about the quality of care being provided. Finally, the participants raised the critical question of whether the vast majority of care that falls above the “inadequate” level, but fails to reach the same levels of quality seen in enriched early intervention programs, is “good enough” for low-income children who are experiencing the stresses and disadvantages associated with living in poverty.

These data focused the workshop participants' attention on the fundamental importance of protecting children from arrangements that are found to be unsafe, unsanitary, and/or entirely unstimulating. Participants also found it valuable to draw a distinction between initiatives aimed at minimizing poor-quality care, such as regulatory reform, and those aimed primarily at moving more mediocre or average arrangements toward the upper end of the quality spectrum, such as provider-training initiatives and accreditation programs.

Having documented wide variation in the quality of available child care, what did the workshops reveal about differential patterns of enrollment in poor-quality and good-quality arrangements by family income? On one hand, a set of constraints appears to limit the options from which low-income working families, particularly those without the day-to-day support of another adult who can share child care responsibilities, can choose (see Chapter 2). On the other hand, perhaps all families face constraints and, although higher-income families spend more on child care, little is known about whether they are purchasing higher-quality care with their additional dollars.

The evidence presented at the workshop suggests that the small share of low-income families who rely on center-based care face a range of quality similar to that of higher-income families. Conversely, low-income families who rely on home-based and informal arrangements may be receiving relatively poorer-quality care. With respect to center-based care, the NICHD Study of Early Child Care, the Cost and Quality Study of center-based care, and the Profile of Child Care Settings study—a nationally representative survey of child care centers and regulated family day care homes (Kisker et al., 1991) —found that the distribution of children across centers of differing quality did not vary by family income. The resource and referral agency staff interviewed by Sharon Long also reported that quality of care for children in subsidized arrangements was as high as that for children in privately funded arrangements. To the extent that other studies have found associations between center quality and family income, it appears that working-class and lower-middle-income families—not families in poverty—are relying on poorer-quality child care centers (Hofferth, 1995; Phillips et al., 1994).

The workshop participants speculated that low-income children may actually get the best and worst of center-based care, depending on their access to highly subsidized intervention and other center-based programs. For example, the U.S. Department of Education's Life in Preschool study (Layzer et al., 1993), presented by Barbara Goodson, found that Head Start and public-school-sponsored prekindergarten programs provided higher-quality care than a sample of community-based child care centers that served comparably low-income preschoolers. Head Start and school-based programs that serve low-income children, as distinct from typical child care centers, are also more likely to provide health services and developmental testing (Hofferth and Kisker, 1994).

This pattern of findings could produce an “average” level of quality for low-income children in centers that marches that of higher-income families while camouflaging inequities within the low-income population. While little systematic information is available regarding who, within the low-income population, gains access to the higher-quality intervention programs, it is known that they tend to operate part-day and even part-year. This makes it very difficult for full-time working low-income parents to avail themselves of these programs (see Hofferth, 1995, for a fuller discussion of these issues).

With respect to other forms of care, however—notably home-based and relative care—both the NICHD study and the Study of Relative and Family Care found that numerous markers of economic and social disadvantage, including measures of family income, predicted enrollment in lower-quality arrangements. The NICHD study found, for example, that families with higher levels of nonmaternal income were able to avail themselves of higher-quality care. It appears that sources of income beyond that provided by the mother, including subsidies, are necessary for low-income families to purchase higher-quality care. To place these findings in context, it is important to reemphasize that low-income children are disproportionately cared for in arrangements other than child care centers. As discussed earlier, however, there is growing interest in—and reliance on—center-based care in this population.

The emerging literature on typical child care for low-income children has raised an additional issue, namely continuity of care, that many workshop participants believe should be part of the discussion of child care quality. The attention of the workshop participants was riveted on this issue by data from the NICHD study showing that, across all income groups, 35 percent of infants had experienced at least three different arrangements—either sequentially or layered on top of each other—by age 12 months. The National Child Care Survey examined simultaneous reliance on multiple nonparental providers and found that about 24 percent of low-income children under age 5—and 45 percent of low-income

preschoolers in families headed by an employed single mother—were in more than one arrangement on a regular basis.

This instability derives from multiple sources: the rapidly shifting employment circumstances of low-income families, unstable child care arrangements, and, in general, the volatile lives of poor and low-income families who are struggling to attain and sustain paid employment. For example, Meyers reported that nearly one-third of children whose mothers enrolled in the GAIN employment and training program were in and out of two or more different child care arrangements over the course of a year, the majority of transitions caused by changes in parents' schedules. Siegel and Loman's Illinois data indicated that more than one in five children were placed with two or more providers during a given week, largely as a result of parents' difficulties matching work and school schedules with the schedules of child care providers.

Substantial evidence has demonstrated that variation in child care quality—within the range available in typical community- and home-based programs—affects a wide range of children's developmental outcomes, including cognitive, social, and health outcomes. In most cases, however, the short-term effects— while significant in a statistical sense—are modest, particularly when the influence of children's home environments on development is also taken into account. Virtually no research has examined the cumulative, long-term effects on children of attending child care arrangements of varying quality as preschoolers or the societal impacts of exposing a substantial share of young children to child care settings that fail to foster their optimal development. Only a handful of very recent studies have observed children in the vast, informal child care market consisting of care by relatives, friends, and neighbors. And this research has not incorporated the high-quality early childhood settings that have been studied in the context of evaluations of early intervention programs. As a result, it is less likely to capture the positive developmental effects that children living in poverty appear to derive from high-quality, comprehensive programs. At the workshop, the focus of interest was on the effects on low-income children of variation in quality of child care.

Results from the Cost and Quality Study provided some preliminary evidence that poor-quality care may have a stronger impact on lower-income children than on their more advantaged peers, particularly with respect to prereading development (Peisner-Feinberg, 1995). The smaller

North Carolina study presented by Burchinal did not, however, find that poverty status of the children moderated the effects of quality on children's development.

Deborah Vandell reported findings regarding after-school care for low-income third graders in Milwaukee (Posner and Vandell, 1994). Although family incomes were lower for children who attended formal after-school programs than for children in either informal adult supervision or self-care, the children in more formal programs had better grades in reading, math, and conduct and were rated by their teachers as having better work habits and peer relations than children who were informally supervised after school. It is notable that children who attended formal after-school programs spent more time engaged in academic activities and enrichment lessons and less time watching television than children in other forms of after-school care.

In sum, the relatively new literature that examines the effects of variation in the quality of typical child care on low-income children 's development is beginning to suggest that quality, structure, and appropriate adult supervision may matter more for low-income children than for their higher-income peers. This is an especially promising area for further research given its significant implications for choices that must invariably be made between using subsidies to expand low-income families' access to the existing supply of care or to upgrade the quality of the care that their children receive. What, for example, are the long-term consequences of modest impacts of poor-quality care on children's social, language, and cognitive development? In addition, very few studies include direct measures of the quality of care provided by mothers at home as part of the same design (the NICHD Study of Early Child Care is a notable exception). What is known about how children's home and child care environments interact to affect development, particularly at the extremes of quality? And what are the effects of variation in the quality and continuity of care on the quality and consistency of childrearing that parents are able to provide? The workshop participants emphasized that efforts to understand the distribution of low-income families across differing types and qualities of child care warrant careful attention, particularly insofar as they are linked to the employment status of low-income parents.

Parents from all income groups indicate that their primary consideration when selecting child care is to find a quality environment for their

children. Some of the ingredients of quality that they consider closely match the factors that research has identified as important, namely, the provider's warmth and children's access to educational opportunities. They do not, however, generally consider the education and training of the provider, nor the regulatory status of the arrangement, as pertinent to quality of care—both factors that research has revealed to be important.

Research presented at the workshop, however, confirmed what prior research has already revealed regarding the contributions that provider training and the aspects of care that are typically regulated (i.e., ratios and group size) make to the nurturance and educational opportunities that children receive in child care. The NICHD Study of Early Child Care, for example, found that the best predictors of higher-quality, home-based child care were group size, ratios, and the specialized training and childrearing attitudes of the provider. These factors predicted the more positive caregiving that parents seek.

Other pertinent evidence from a longitudinal study designed to examine the effects of an improvement in Florida's child care regulations on children's development¹ was reported by Carollee Howes. Children enrolled in child care centers after the improvements had been made, compared with those in the same centers prior to the changes, showed significantly more advanced cognitive, language, and social development. They were also observed to be more securely attached to their teachers and were rated as having fewer behavior problems.

The share of subsidized children served in the centers did not change during this period. What changed dramatically was the share of classroom teachers who had received specialized training, as required by the new regulations, as well as the percentage of classrooms that were observed to offer good-quality care (from 25 to 36 percent for infants and toddlers and from 27 to 44 percent for preschoolers). The teachers ' sensitivity and responsiveness were observed to increase and negative management styles were observed to decline from pre- to postregulatory improvement.

The workshop participants discussed the implications of the apparent gap between parents' perceptions and the empirical evidence regarding

the important inputs for a quality child care environment. It is important to determine whether consumers are even knowledgeable about the research evidence on these issues. In this context, two major concerns were raised. First, absent consumer pressure to ensure that more child care providers offer children the inputs that research shows to be important, there is little incentive for providers to offer such inputs or for policy makers to expand access to these features of care. Second, there is a tension between an interest that some have in educating consumers about the value of regulated dimensions of care and of provider training, and a concern that, absent access to arrangements that correspond to research-based perceptions of good care, consumer education may foster anxiety. Others noted, however, that educated and perhaps anxious consumers are most likely to call for changes in the child care market.