the literature, as discussed in the next chapter, calls attention to the myriad ways in which children's relationships with their parents (or those who otherwise serve as the child's primary caregivers), the parents' behavior toward their children, and the home environment in which children grow up profoundly affect what children learn and can do, what they expect and believe, and how they approach others during the early years and start them off along differing pathways as they move into the school-age years.

Arguably, young children now growing up in the United States are exposed to an unprecedented number and variety of out-of-home environments. School entry used to mark a major transition when the balance of a child's time spent at home and with parents was profoundly altered. Today this happens for the majority of children before the end of their first year, given trends in parental employment and early reliance on child care. It is certainly plausible that, as a result, adults other than parents, care settings other than the child's home, and peers and neighborhood settings are becoming increasingly influential sources of early developmental variation. We do not yet, however, have any evidence bearing on this speculation. In fact, studies of both child care settings and neighborhoods have reaffirmed the powerful influence of the family and studies of socioeconomic influences have emphasized the large extent to which they affect young children through effects on their caregivers.

While there has been a long-standing agreement among those who study children that development cannot be understood out of context—the so-called ecology of human development (Bronfenbrenner, 1979, 1986) —concerted efforts to understand influences that derive from contexts other than the immediate family are relatively new. These efforts include studies of children as they grow up in families that occupy different socioeconomic niches, experience nonparental child care, and reside in communities and neighborhoods with widely differing characteristics and resources. Research that tracks the natural trajectories of young children, particularly longitudinal studies that follow the same children over time, tell us about how these environments affect the natural unfolding of development. Do these beyond-the-family contexts matter, and if so, how much do they matter in shaping the early direction of children's lives? We also learn about contextual influences from studies of efforts to change these environments, ranging from providing infants and toddlers with enriched child care to moving families out of dangerous neighborhoods. This research tells us about the malleability of early development. Can we change development by changing its contexts, and what does it take? We discuss both of these streams of research and argue for their integration.

Research on the context for early development that is provided by parents and other primary caregivers in the home (Chapter 9) provides the point of departure. We then summarize the research on socioeconomic

influences on early development (Chapter 10), including the influence of income and poverty, parental schooling and employment, and family structure. Next, we consider the influence of child care (Chapter 11). Although we introduce the literature on child care in Chapter 9, particularly as it bears on young children's attachments to important caregivers, we dedicate a separate chapter to examining child care as an important setting that supplements the care children receive from their parents. Nevertheless, as we illustrate, families blend and switch among various forms of exclusive parental care and care that is shared with others, making distinctions between these two contexts fuzzy at best. We close Part III with a discussion of neighborhood influences on development (Chapter 12).

There is one very important context for early development that is not addressed in this report, namely the media. Today's children spend more time with more media (e.g., television, VCRs, CD players, game systems, computers, among others) than any generation before them, and there is every reason to believe that their media use and exposure will continue to increase (Roberts et al., 1999). Children ages 2 to 4 spend well over 4 hours every day, on average, exposed to the media (primarily television) and over a quarter have televisions in their bedrooms. In many instances, these are noninteractive experiences. For example, almost 15 percent of 2- to 7-year-olds watch television mainly alone (Roberts et al., 1999). We are only beginning to understand the repercussions of these trends for family life and child well-being. Our neglect of this topic is not a signal of any lack of concern; this is clearly an issue that warrants substantial attention.¹

The science reviewed in this part of the report is more interdisciplinary than that reviewed in Part II, particularly as it moves beyond parents and the home environment. Efforts to understand the effects on development of economic influences, and notably of poverty, have brought together the full spectrum of social and behavioral scientists. Child care has been studied primarily by developmental psychologists, but also by sociologists and economists. As a result, the science base undergirding the understanding of contextual influences has benefited from different methodologies, theoretical perspectives, and standards of evidence, making it both rich and full of controversy.

This literature is not without its shortcomings, however. The major challenge facing those who seek to understand how beyond-the-family contexts affect early development is that parents select these environments. They decide where to live, where and how much to work, whether and

when to place their babies in child care and which child care settings to use, and how to invest their resources. Thus, effects on children that are ascribed to such factors as child care and neighborhoods may, in fact, really be effects of parent selection.

This distinction is extremely important for policy purposes. If, for example, quality child care is associated with children's development because parents who also provide more for their children at home place their children in higher-quality programs, then efforts to improve quality will produce smaller improvements in child well-being than anticipated. If this were true, a more effective strategy would be to provide family-based benefits to more children. Researchers deal with this problem in two ways. First, they measure the family environment and control for it statistically when examining the effects of nonfamily environments, effectively measuring the effects of child care, for example, net of at least some of the effects of the family and home environment. These controls, however, can never capture all of the family environment. Second, they also conduct experiments in which children are more or less randomly assigned to different child care programs (as is the case with early intervention studies) or to different neighborhoods. Of course, children cannot be assigned to different social classes, but one can measure the effects of changes in socioeconomic status (SES) and the effects of SES on siblings who experienced different socioeconomic contexts during the early years of life.

The second challenge concerns genetic influences. Behavior geneticists, in particular, have argued that parents exert their influence on children primarily through their genetic contributions (Rowe, 1994), yet it is increasingly clear that the expression of heritable traits depends substantially on experience, including how parents behave and what they provide for their children (Collins et al., 2000). Behavioral geneticists also argue that socioeconomic status and the benefits it confers on children (including the neighborhoods they live in) largely reflect parents' genetic endowments. As such, efforts to relate children's family SES to their achievement, for example, that do not somehow adjust for parent and child genetic endowments risk overattributing to SES causation what ought to be attributed to genetic influences. Although this argument has merit, we discuss research that places genetic factors in the context of substantially larger SES-based environmental influences on children's development.

With these challenges in mind, our synthesis of research on family-based caregiving influences, socioeconomic influences, child care, and neighborhood influences on early development focuses on the role of experience in early development. The themes that emerge reflect several presented in previous chapters:

• There is a firmer understanding of what constitutes “deprivation,” in contrast to “enrichment,” during the early years. At the most general level, it is clusters of influences that undermine development rather than isolated or temporary exposures to harm. Far less is known about how bad or enduring these influences need to be in order for change to be done, about factors that enable some children exposed to extremely detrimental circumstances to avoid harmful consequences while others succumb to serious problems, and about capacities and opportunities for recovery.

• At the other end of the spectrum, beyond numerous associations between better environments and better development, little is known about what constitutes “good enough” environments and how the answer to this question varies for different children and families. Similarly, enrichment remains an elusive construct, particularly insofar as it is conceptually linked to contemporary interest in accelerating or improving the development of children who are relatively risk free and on track.

• While there is no evidence that critical or sensitive periods characterize parenting, socioeconomic, child care, or neighborhood influences, there is suggestive evidence that young children compared with older children may be particularly vulnerable to very detrimental experiences that derive from aberrant caregiving and serious economic hardship.

• Children's early pathways can be shifted by efforts to change the contexts of their lives. Less is known, however, about what it really takes to shift the odds, and very little is known about the factors that keep children moving along adaptive pathways once they leave the early childhood years behind.

In this report, we have emphasized the remarkable achievements of young children and the strong developmental thrust that characterizes these accomplishments. Each achievement—language and learning, social development, the emergence of self-regulation—occurs in the context of close relationships with others. These close relationships are typically with parents or those who serve the parenting role in the child's life. We turn our attention to these relationships in this chapter and explore their influence on early development. Although we address young children's relationships with their child care providers given their pervasive and significant role in the earliest years of life, we reserve the fuller discussion of child care for Chapter 11. We start the discussion of parenting with a focused synthesis of the extensive literature on parent-infant attachment, followed by a discussion of other aspects of parenting that extend beyond the provision of emotional security. Next, we examine disruptions in parenting. We include an examination of the orphanage-to-adoption research as a demonstration of the extent to which young children can recover from early adverse experiences when the contexts of their lives change dramatically. We use this as a backdrop for an initial discussion of parenting interventions, which is placed in a broader context in Chapter 13.