Diversity in Caregiving Contexts Research Paper

A generation ago, infants and toddlers were being described as the new school-age children, in response to what then was an unprecedented demand in the United States for infant and toddler child care (Fitzgerald, Strommen, & McKinney, 1977). In 1940 only 10% of children had mothers in the labor force. By 1970 that percentage had risen to 36%, and today nearly 70% of all children under 5 years of age have mothers who work outside the home (Hernandez, 1997). Impetus for the accelerated demand for child care of very young children that occurred from 1940 to 1970 was fueled by many factors, not the least of which was Franklin Delano Roosevelt’s authorization of the Works Progress Administration toward the end of the Depression, which established the first U.S. government-funded child care programs for families in need. In 1944 Congress passed the Lanham Act, thereby authorizing use of federal funds to support child care for all children, beginning at 2 years of age. By 1945 there were slightly more than 1.5 million children enrolled in full day care centers or nursery schools.

Reviews of the child care literature have described historical trends in the study of nonmaternal child care (Scarr & Eisenberg, 1993; Clarke-Stewart, Allhusen, & Clements, 1995), cross-cultural perspectives on siblings as caregivers (Zukow-Goldring, 1995), kinship care (Wilson & Chipaungu, 1996), contextual setting events and cross-cultural variations in child care (Lamb, 1998), and criteria for selecting child care (Honig, 1995). The extant literature on infant and toddler child care supports the broad conclusion that supplemental nonmaternal child care is not harmful to infants and toddlers—and in fact can be beneficial both to child development and to the quality of the parent-child relationship. Nevertheless, there are many issues that remain unresolved about supplemental child care during the birth-to-three years. There is consensus that quality is the key to success of child care, regardless of where that care takes place. However, the defining features of quality have not been fully articulated— either for parental child rearing or for nonparental child care. There is increasing recognition that fathers play an important role in early child development, but there is not much specificity as to what that role actually is. We know far more about characteristics of children that are correlated with father absence, but we know little about the effects of father presence on child development. Although there is considerable research assessing the impact of nonparental child care on child development, no theoretical or conceptual model currently provides an overarching framework to guide systematic research in this area. Attention is directed to issues related to the physical settings within which nonparental child care takes place, with a distinct focus on health and safety issues, child-to-adult ratios, and training of care providers. In contrast, less attention is given to issues related to the quality of the care provider-child relationship. If the quality of the adult-infant relationship comprises the core of human socialemotional development as attachment theorists would have it, then one would expect that the relationships that infants and toddlers have with their child care providers would contribute to emotion regulation during early childhood.

In this research paper we examine selected aspects of the child care literature to identify streams of research that may best inform efforts to establish benchmarks for quality care of infants and toddlers, particularly in nonparental child care settings. We propose that the study of the impact of early child care would be facilitated by adopting ecological approaches—such as that embodied in systems theory—in order to provide organizational structure to the field. We briefly review several national and local studies of the impact of early child care that have in common use of randomized designs and research designs that provided tracking of children over time. Then we draw attention to four focal issues— infant temperament, caregiver-child relationships, father involvement, family risk load—and suggest that each should be a targeted domain in the overall assessment of the impact of nonparental child care on infant and toddler development. Finally, we close with a discussion of evidence that may be used to establish benchmarks of quality for nonparental care of infants and toddlers.

Nonparental Child Care of Infants and Toddlers

Today’s demand for supplemental child care for infants and toddlers makes yesterday’s reference to new school-age children a gross understatement. According to data from the National Household Survey (1999), 61% of all children 4 years of age or under participated in some form of regularly scheduled child care; this includes 44% of infants younger than 1 year of age, 53% of 1-year-olds, and 57% of 2-yearolds. Today, approximately 12 million children younger than 4 years of age receive some form of partial or full-time nonmaternal child care involving as many as 40 or more hours in nonparental care.

Throughout human history, the care of children rarely has been the mother’s exclusive responsibility. However, powerful theories of human development emerged during the twentieth century that focused attention to mother as the primary caregiver for infants and toddlers. Early in that century, Freud’s (1946) psychoanalytic theory and Piaget’s (1952) theory of genetic epistemology drew attention to the first 3 years of life as uniquely important for personality development and cognitive development, respectively. Bowlby (1969) drew heavily from psychoanalytic theory to articulate an extraordinarily influential theory of early personality development as emergent from the quality of the mother-infant relationship. Prior to Bowlby’s work, Spitz (1965) focused attention on the impact of maternal and sensory deprivation on infant development. Ainsworth subsequently developed a research procedure, the Strange Situation, that has had a profound impact on the study of the quality of the mother-infant social-emotional relationship (Ainsworth, Bleher, Waters, & Wall, 1978). Although attachment theory and research methods have markedly advanced knowledge of early personality development, affect regulation, and interpersonal relationships, they also created an atmosphere highly suspicious of nonmaternal child care, especially during the first year of life. In addition, they directed such strong attention to the motherchild relationship that the father nearly became a forgotten parent with respect to his parenting role during early childhood development (Coley, 2001). Although few developmental theorists continue to hold onto the view that the events of infancy rigidly determine subsequent outcomes (Rutter, 1995), contemporary brain research asserts that the impact of environmental events on neurobehavioral systems in the early years of development need to be carefully considered (Shonkoff & Phillips, 2001).

There is a tendency in the literature to discuss care of children by individuals other than a parent as nonmaternal child care. Although many authors have drawn attention to the fact that throughout human history, “the exclusive care of infants by their mother is an exception rather than a rule” (Werner, 1988, p. 105), supplemental caregivers tend to be other women more often than to be men. Even in cultures that have official policies granting parental leave to men, relatively few men participate equally in the routine care of their infants and toddlers (Mackey, 1996; Salmi, 1994). For example, Finland has one of the highest percentages of women in the workforce of any country of in the world (80%). Parents receive 11 months of parental leave, the first 105 days of which are reserved for mothers. Thereafter, mothers and fathers can choose among a variety of options for allocating the remaining months of parental leave to one or the other parent. However, in relatively few families (4%) do fathers choose to take advantage of the parental leave policy (Salmi, 1994). After 11 postnatal months, families may enroll their toddlers in municipal day care (government-supported child care centers), private day care, or at-home day care. Families choosing options other than municipal child care receive government subsidies to help offset the costs. In Finland (as elsewhere) parental leave policies are associated with lower demand for nonparental infant child care (Kamerman & Kahn, 1995).

The options for nonparental child care in Finland are essentially those found everywhere. Nonparental child care basically takes place in one of two basic settings: in a center or in someone’s home. Centers can be located practically anywhere—schools, workplaces, religious institutions, universities, shopping centers, apartment buildings, or independent private buildings. Although center-based care historically involved preschool-aged children, the number of infants and toddlers in such care is increasing. Home-based child care varies, with some infants and toddlers cared for in their own homes (at-home day care) and others cared for in someone else’s home, often referred to as family day care or kinship care (Wilson & Chipungu, 1996). Child care that is provided in the child’s own home is convenient, but it is also the most expensive type of child care when it is provided by a nonrelative. Kinship care that takes place in the child’s own home generally is provided by grandparents, older siblings, or others with the likelihood that services are free or of minimal cost (Wilson & Chipungu, 1996). On the other hand, kinship care often involves low-income families and chosen because it is the only affordable option available to families. When kinship care takes place in the home of a relative, the financial arrangements vary greatly. The issue of financial support to enhance the quality of kinship care is a matter of current policy debate (Hornby, Zeller, & Karraker, 1996). Because kinship care and family care take place in the child care provider’s home, it often means that the number of children being cared for is relatively small. When several children are involved, their ages may vary substantially. Multiple age groupings retain a sense of family structure, with older children available as models and teachers for younger children.

Traditional caregiving provided by nonparental persons to infants and toddlers while their parents work is referred to as child care. Nonparental caregivers are referred to as child caregivers, child care providers or workers, domestic caregivers, educators (those providing both education and care), or teachers. Although some may also be called day care workers, the term day care is used less often as more parents work nontraditional hours, thus requiring child care services evenings, overnight, and on weekends. The term child care provider is rapidly becoming the generic descriptor for all child caregiving services provided to infants and toddlers in the absence of their parents. Regardless of the descriptor used to depict nonparental care providers and regardless of where such care takes place, there is a need for some overarching organizational structure to guide the systematic study of the impact of nonparental care on infant and toddler development in all of its cultural complexity and diversity (Cochran, 1993; Kontos, Howes, Shinn, & Galinsky, 1995; Morelli & Verhoef, 1999). For example, studies of the impact of nonparental child care on child development does not yet take into account the full impact of such ecological variables as geographic region, family structure, and maternal characteristics (Singer, Fuller, Keiley, & Wolf, 1998). Singer et al. demonstrated that the type of child care parents choose is influenced by such variables as parental race and ethnicity, geographic location of residence, maternal employment, child age, maternal education, and number of children in the family. We believe that ecological models of human development—especially those that embrace some variant of systems theory—provide the best hope for organizing and directing the systematic study of such diverse sources of impact.

Systems Theory: An Organizing Framework for The Study of Nonparental Child Care

In the 1930s Ludwig von Bertalanffy, dissatisfied with both mechanistic and vitalist attempts to explain the organization of living things, articulated an alternative organismic position which he subsequently named general systems theory (von Bertalanffy, 1950, 1968). Since then a number of investigators have expanded and refined systems theory into a powerful conceptual framework for organizing the study of adaptive behavior and adaptive functioning applicable across the life span. Systems theorists view the organization of all systems as emergent, epigenetic, constructive, hierarchically integrated (Bronfenbrenner, 1979; Ford & Lerner, 1992; Gottlieb, 1991; Miller, 1978; Sameroff, 2000) and potentially chaotic (Thelen & Smith, 1994). Moreover, systems theorists believe that the ecological context within which systems develop plays a key role in the organizational process. Because the components of any system are interdependent, it follows that attempts to understand adaptive functioning at one level require that we understand how that level is embedded within the broader system of which it is a part. Moreover, because systems theory posits that all components of a system are interdependent, the parts derive their meaning only in the context of intersystem and intrasystem relationships.

Assessing the impact of any kind of child care on child development requires longitudinal research approaches. One can easily assess proximal impacts with cross-sectional designs, but only through longitudinal designs can one assess changes that occur in relation to events that occur over the life course. From a systems perspective, identifying causal factors related to child care outcomes must take into account intra-individual (within the individual), interindividual (between individuals), contextual (social-historical-temporal events or situations), and organism-environment transaction (ecological, bidirectional) sources of variance. Relying on main effect or direct effect models is unlikely to reveal much about the dynamic changes that occur during the birth-to-three years.

Within this multilevel approach, we focus on four aspects of the child care system: primary system characteristics (characteristics of individual units—child, family, neighborhood—of a system), intrasystem relationships (parent-child, spousal, sibling, kinship interrelationships), adjunctive system influences (ecological contexts that impact the primary system), and intersystem relationships (boundaries, barriers, transitions; Fitzgerald, Zucker, & Yang, 1995). Recent accounts of the impact of neighborhood poverty on child development illustrate the nesting of risk factors in communities (Duncan & Brooks-Gunn, 1997) and support the hypothesis that risky neighborhoods envelop individuals and families within maintenance structures that sustain risk (Zucker et al., 2000). For example, very young children reared in antisocial alcoholic families are more likely to be exposed to parents who themselves had childhood behavior problems, illegal behavior, frequent arrests, chronic lying, relationship disturbances, depression and family violence, neuroticism, poor achievement and cognitive functioning, and low socioeconomic status (SES; Fitzgerald, Puttler, Mun, & Zucker 2000). There is also evidence of assortative mating in such families. This suggests that the nesting environment within which children with a high family history for alcohol are reared also carries a substantially higher risk load (Zucker et al, 2000). Thus, antisocial alcoholic parents provide a high-risk rearing environment that is very likely to be embedded within a highrisk neighborhood (cf. Osofsky & Fenichel, 1994).

All development takes place in a complex environment that consists of the primary system, the transactions that occur within the primary system (Sameroff, 2000), and the transactions that take place between the primary system and all of its adjunctive systems (Carlson & Cassell, 1984; Fitzgerald et al., 1995). Contemporary prevention programs designed to enhance child development during the early years reflect this thinking. They address issues related to child development, parent involvement, consistency of care, and networking to the broader community of human service agencies and to the schools. To the extent that the primary system is embedded in risk, adjunctive child care systems may be the child’s best hope for being exposed to individuals and environments that will stretch the boundaries that encapsulate the primary system and will thereby generate resilience structures. The question of specific interest for the current research paper is whether nonparental child care generates resilience, enhances risk, or at the least does no harm. With the exception of large federally funded studies of the impact of child care programs for primarily low-income families, answers to this question will require use of large-representative-sample, multifactorial, interdisciplinary, prospective longitudinal research designs more extensive than has typically characterized research attempting to determine the impact of nonparental child care for infants and toddlers.

Early Child Care and Child Development

During the last third of the twentieth century, many studies of child care programs appeared in the scientific and professional literatures. Often these studies described programs that were narrowly focused—guided by a particular theoretical model. Rarely did studies involve random assignment of participants, use sophisticated quasi-experimental designs, or collect outcome data over time. Moreover, they seldom involved sample sizes sufficient to generate reasonable statistical power. Near the end of the century, however, significant funding from several government agencies established largescale, longitudinal evaluations of the impact of early child care programs. Nearly all of these studies emphasized child development, parent involvement, and community networking; some utilized random assignment evaluation designs, some were center based, some were home based, and many combined home- and center-based models. Although richly diverse, modestly successful, and collectively limited, when combined with the extant literature these studies provide support for the positive effects of quality supplemental care on early child development, family functioning, and community networking.

During the last decade of the twentieth century, agencies of the United States government provided funding for significant evaluations of the effectiveness of child care programs that involved random assignment, large sample sizes, and repeated assessments over time. For illustrative purposes we describe the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care, theAdministration for Children,Youth, and Families national evaluation of Early Head Start, and the Parent-Child Development Centers (which actually were established as long ago as 1969–1970). In addition, we briefly review outcomes from the Carolina Abecedarian Project, for which there are substantive longitudinal findings. For our final example, we describe outcomes from the Goteborg (Sweden) Child Care Study.

NICHD Study of Early Child Care

The NICHD Study of Early Child Care was designed to assess the impact of nonparental caregivers on a broad range of child outcomes, including social-emotional development (interpersonal relationships, self-regulation), cognitive development (reasoning and problem solving), linguistic development (receptive and productive language), achievement performance (literacy, numeracy, school readiness), physical development (height and weight), and health (immunizations, chronic illness, illnesses of childhood). Nonparental caregivers included relatives (kinship care) and nonrelatives. The context for caregiving varied from in-home, in the provider’s home, and at child care centers. The NICHD study was guided by ecological (Bronfenbrenner, 1979; Bronfenbrenner & Morris, 1998) and developmental lifecourse theoretical perspectives. Thus, the study was designed at the outset to examine the impact of contextual influences as well as the interaction between context and age-related experience. Some experiences are normative in that they occur at narrowly circumscribed ages for most people (e.g., school entry). Other experiences are nonnormative in the sense that they are not linked to specific ages (e.g., parental divorce, change in child care arrangement, family relocation). When combined, the ecological and life course perspectives provide a framework for conceptualizing the relations of both normative and individual contextual influences to developmental pathways.

Because the NICHD study is longitudinal, it offers the possibility of tracking the effectiveness of different models (i.e., cumulation, endurance, sleeper, fade) proposed to account for the way in which early care experiences influence later child behavior. The cumulation model suggests that child care contributions aggregate over time such that children in child care should show progressively stronger effects over the course of their exposure. The endurance model proposes that the effects of child care consolidate and persist over time regardless of changes in the child’s educational context. The sleeper model predicts that child care effects are least likely to be evident during the time that infants and toddlers are enrolled in child care, but emerge upstream at later points in developmental time. The fade model suggests that the effects of child care are transient and so will disappear over time as more proximal events come to exert their influences on child behavior. The NICHD not only provided an opportunity to assess these models, but also allowed investigators to examine the relations between child care experiences and concurrent psychological and health outcomes, the effects of the home environment on child outcomes, and the linkages between demographic and family characteristics and child development (NICHD Early Child Care Research Network, 1994, 1996, 1997, 2000).

Participants in the study were recruited from 24 designated hospitals at 10 data collection sites across the United States. Factors such as location, availability, previous working relations with the site investigators, and the nature of the patient load contributed to the selection of hospitals within sites. A total of 1,364 newborn infants (and their families) were enrolled in the study, with the sample distributed approximately equally over the 10 sites. The enrolled families included mothers who planned to work full time (53%), part time (23%), and not at all (24%) during the child’s first year. The enrolled families came from a wide range of socioeconomic and sociocultural backgrounds, as well as from diverse family arrangements (24% ethnic minority children, 11% mothers who did not complete high school, 14% single mothers; these percentages are not mutually exclusive).

On a weekly basis, each site was expected to screen a minimum of 20 newborn infant-mother dyads in the participating hospitals for potential enrollment to the study. The exclusion criteria for the hospital screening included (a) maternal age (˂18years), (b) language proficiency (non-English speakers), (c) family mobility (family planned a move with 12 months), (d) infant medical status (medical complications or maternal substanceabuse),(e)maternalillness,(f)adoptionplacement, (g) lack of maternal cooperation in data collection, (h) family involvement in other research, (i) residence in an neighborhood posing excessive danger for data collectors, (j) maternal failure to complete the hospital interview, and (k) other factors. In addition, researchers collected information about the child’s gender, gestational age and weight, the mother’s ethnic and racial identification, age, education, employment status, and her partner’s residential status and education. Of the original 1,364 families, 1,100 were still participating in the study when most of the children were entering the third grade, yielding an 80.6% retention across the 8 years of study involvement.

Supplemental funding provided by the U. S. Department of Health and Human Services enabled investigators to add direct measures of fathers’ attitudes and perceptions in 6 of the 10 sites (Arkansas, California, Kansas, North Carolina, Pittsburgh, and Wisconsin). The fathers’ component enabled investigators to examine not only fathers’ direct impact on their children, but also the impact of fathers on the quality of the marital relationship and the impact of maternal employment. Experience during early development was assessed through a diverse array of measures designed to capture the child’s experience in the context of home and family, in child care, and eventually in school. Measures of socialemotional, cognitive, linguistic, and academic development and physical growth and health were used to assess children’s developmental status. Selection of measures was based on (a) the child’s developmental level, (b) the psychometric properties of the measure, (c) the applicability of measures to children and families varying in ethnicity and socioeconomic status, (d) the amount of time needed to complete the measure, (e) the relations among the different measures planned for each visit, and (f) the results of pilot testing. Two criteria were considered in selecting specific child outcomes to be assessed: (a) that the developmental importance of the outcome construct was well documented in previous research and theory, and (b) that there was reason to hypothesize that children’s development in a particular domain would be affected by environments of early child rearing.

What have investigators found to date? Reports from the National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (2000) suggest that during the children’s first year of life, there was high reliance on infant care, very rapid entry into care postbirth, and substantial instability in care. By 12 months of age, 84% of the infants in the study had entered some form of nonparental child care, with the majority starting care before the age of 4 months. When they first entered care, 25% of the infants were cared for by their father or their mother’s partner, 23% were cared for by other relatives, and only 12% were enrolled in child care centers. Over the first year of life, the majority of children in nonparental care experienced more than two different child care arrangements, and more than one third experienced three or more arrangements.

Economic factors were most consistently associated with the amount and nature of nonmaternal care infants received. For example, mothers with higher incomes and families that were more dependent on the mother’s income placed their infants in child care at earlier ages. Maternal personality and beliefs about maternal employment also contributed. For example, mothers who believed that maternal employment has positive effects on children put their children in nonmaternal care for more hours. Poor families were less likely than were affluent families to use child care, but poor children who were in care averaged as many hours as did children from other income groups.

Observations of the quality of care at 6 months indicated that more positive caregiving occurred when children were in smaller groups, child-adult ratios were lower, caregivers held less authoritarian beliefs about child rearing, and physical environments were safe, clean, and stimulating. Observed quality of care for poor children was generally lower than that for nonpoor children when they were cared for by an unrelated caregiver. The single exception was that poor children in centers received better-quality care than near-poor children, perhaps because they were more likely to be in subsidized (and therefore perhaps more regulated) settings. Evaluation of child care centers in relation to guidelines recommended by professional organizations for child-staff ratios, group sizes, teacher training, and teacher education indicated that most classes observed in the study did not meet all four of these guidelines.

Analyses of the effects of family and child care on child outcomes indicated that in general, family characteristics and the quality of the mother’s relationship with her child were stronger predictors of child outcomes than were child care factors. Family factors predicted child outcomes even for children who spent many hours in child care, and statistically significant child care effects were relatively small in size. Maternal depressive symptoms comprised one family predictor of child outcomes (in addition to income level, education, attitudes, and behavior). Children whose mothers reported feeling depressed performed more poorly on measures of cognitive-linguistic functioning at 36 months and were rated as less cooperative and more problematic. However, depression effects on expressive language and ratings of cooperation were moderated by maternal sensitivity, with sensitivity predicting better outcomes more strongly among children of depressed mothers.

Analyses controlling for nonrandom use of child care by families of different socioeconomic backgrounds revealed that among the aspects of child care studied, a relatively consistent predictor of child outcomes was the observed quality of care. When observed quality of caregivers’ behavior was high, children had better cognitive and linguistic abilities, showed more cooperative behavior with mothers during play, and had fewer behavior problems. For children in center care at 36 months, children had fewer behavior problems and higher scores on language comprehension and school readiness when classes met more of the guidelines recommended by experts for ratios, group sizes, and teacher training and education. Higher-quality child care was also associated with higher-quality mother-child interaction among the families that used nonparental care. Additionally, poor-quality child care was related to an increased incidence of infant insecure attachment to mothers at 15 months, but only when the mother was also relatively low in sensitivity and responsiveness.

Overall, type of child care by itself appeared to have relatively limited impacts on child outcomes. At age 3, greater cumulative experience in center care and early experience in child care homes were both associated with better performance on cognitive and language measures than were other forms of care, assuming comparable quality of caregiving environment. Experience with group care (settings with at least three other children, not counting siblings), whether in centers or child care homes, made some difference in several social-emotional outcomes at ages 2 and 3. Children with more cumulative experience in group care showed more cooperation with their mothers in the laboratory at age 2, less negative laboratory interaction with their mothers at age 3, and fewer caregiver-reported behavior problems at both ages. However, higher amounts of group experience before 12 months were associated with more mother-reported behavior problems at age 3, suggesting that benefits from group care may begin in the second year of life.

The quantity of nonparental care was also a statistically significant predictor of some child outcomes. When children spent more hours in child care, mothers were less sensitive in their interactions with their children (at 6, 15, 24, and 36 months) and children were less positively engaged with their mother (at 15, 24, and 36 months, the ages at which child engagement was assessed). In addition, analyses of attachment at 15 months show that children who spent more hours in child care and had mothers who were relatively insensitive and unresponsive were at heightened risk for insecure mother-infant attachments.

ACYF Study of Early Head Start

In 1994, the United States Congress reauthorized the Head Start Act and stipulated that 4% of the appropriation was to be used to establish Early Head Start. The percentage of funding for Head Start diverted to Early Head Start has increased steadily since 1994, and in 2002 reaches its final authorized level of 10% of the Head Start budget (Raikes & Love, 2002). Designated primarily as a child care program for low-income families, Early Head Start is designed to enhance developmental outcomes for infants and toddlers and to enhance family functioning (Jerald, 2000). The 1994 reauthorization of Head Start stipulated that procedures be established to provide for continuous quality improvement of programs and to monitor the fit between programs and community needs. The reauthorization bill of 1998 affirmed both the continuous quality improvement policy as well as the need to evaluate the extent to which Head Start and Early Head Start were achieving their objectives (Raikes & Love, 2002). In order to monitor program quality, performance standards were adopted in 1996. The standards define the services that must be provided to children and families, although they do not specify how such services must be delivered. The lack of specificity for service delivery allows for greater diversity in programming across communities; it also gives programs greater flexibility to develop community partnerships that are unique to family needs for high-quality child care (Fenichel & Mann, 2001).

The Secretary of the Department of Health and Human Services appointed an Advisory Committee on Services for Families With Infants and Toddlers (1994), which identified four key cornerstones of program quality that must be addressed by every Head Start and Early Head Start program: child development, family development, community development, and staff development (see Administration on Children, Youth, and Families, 2001; Fenichel & Mann, 2001). Of the four cornerstones, issues related to child development were clearly identified as first among equals (Fenichel & Mann, 2001). Moreover, the performance standards clearly specify that programs must acknowledge the importance of relationships for infant and toddler development, and this includes relationships with all caregivers, including parents. Because parents are considered to be integral to the delivery of quality child care, Early Head Start programs must address issues related to parenting skills, parentchild relationships, and father involvement. In addition, all children must be screened for developmental disabilities within 45 days of enrollment. Ten percent of all openings in EHS must be available for infants and toddlers with defined disabilities. The number of infants and toddlers with established conditions, environmental risks, or combinations of these is substantial, and current EHS funding is insufficient to provide child care programs for all eligible infants and toddlers because of special needs. To achieve the flexibility necessary to fit programming with community needs (Raikes, Kisker, Paulsell, & Love, 2000), Early Head Start child care program services are offered in formats that are designated as (a) home-based, (b) center-based, (c) home-based, centerbased combinations, or (d) locally designated options. Currently, there are 635 Early Head Start programs nationwide, serving 45,000 infants and toddlers.

Evaluation of Early Head Start

The Administration for Children, Youth, and Families (ACYF) evaluation of Early Head Start (EHS) involves a partnership among ACYF, Mathematical Policy Research, Columbia University Center for Children and Families at Teachers College, 15 universities, and program staff at 17 sites across the United States. Sixteen sites partnered with university research teams and one site was directly affiliated with Mathematical Policy Research, which also served as the evaluation consultant for the full national study. Each site was also independently funded to conduct local studies. Thus, the evaluation of the impact of Early Head Start consists of analyses of the integrated data from 17 regionally distributed programs (Commissioner’s Office of Research and Evaluation, 1999a, 1999b), as well as the independent findings from each of the program sites. In addition, separate studies were conducted to assess the impact of EHS on fathers, child care, health and disabilities, and welfare reform, and a longitudinal follow-up study is currently assessing children as they make the transition from preschool to kindergarten. A similar longitudinal follow-up to track children’s transitions to elementary school is being planned.

The EHS research sites were selected to represent geographical and ethnic diversity within the constraints of the income requirements for eligibility. Across the 17 sites, 3,001 families were randomly assigned to either an EHS program or some other child care program that was available in the local community. Randomization resulted in final samples of 1,513 EHS eligible families enrolled in EHS programs and 1,488 EHS eligible families assigned to the control condition. Federal poverty guidelines were used to determine income eligibility for Early Head Start. Families in the control group were able to secure any kind of child care available in their community with the exception of Early Head Start. They were notified that they would have access to Head Start if they continued to be eligible for the program when their child was 4 years old. Originally, seven of the EHS sites were home-based programs, four were centerbased programs, and six were mixed programs. In the national evaluation, all programs that were not exclusively home-based or center-based were classified as mixed programs. Center-based programs provided all services to families through the center (which also included a small number of home visits). Home-based programs provided EHS services through home visits, although families were able to use other kinds of child care options as they wished. Mixed programs provided services to some families through centers, home visiting, or some combination of these in addition to local options (Commissioner’s Office for Research and Evaluation, 1999a). Initially (1995, 1996), there were five home-based, five center-based, and seven mixed programs. Slightly more than 1 year later, eight programs were homebased, four were center-based, and five were mixed; and by 1999, there were 2 home-based, 4 center-based, and 11 mixed programs (Commissioner’s Office for Research and Evaluation, 2001a). As parental needs for nonparental child care increased, EHS programs responded by providing a greater mix of options for parents. Obviously, the evaluation of program effectiveness will have to take into account changes in the type of child care options available to families over the course of the longitudinal evaluation. Regardless of program type, all programs were required to provide child development services, to build family and community partnerships, and to enlist support staff to provide high-quality services for infants, toddlers, and their families. Thus, all EHS programs are conceptualized as seamless systems, with responsibilities for comprehensive child development, family functioning, relationships with program staff, and connections to community resources.

Evaluation data were collected at enrollment (baseline) and at 5, 15, and 26 months after enrollment (parent services interviews) and when children were 14, 24, and 36 months old (parent interviews, child assessments, and videotaped parent-infant interactions). Only summary data from several aspects of the 6–24 month assessment periods has been completed (Commissioner’s Office of Research and Evaluation, 2001a, 2001b). Although the findings are only preliminary, they suggest that EHS programs have modest positive influences on child development, particularly with respect to enhancement of cognitive and language skills and reduction of aggression and other behavior problems. No differences were found for toddlers’ task-related emotion regulation or their general interactions with mothers during a videotaped freeplay interaction. EHS parents were more knowledgeable about infant-toddler development; engaged in more cognitive, language, and literacy activities at home; were generally more positively engaged with their children during free-play interactions; were less likely to have spanked their child (in the past week); and reported less marital conflict than did mothers in the control group (Commissioner’s Office of Research and Evaluation, 2001c).

Center-based programs had their greatest impact on indicators of child development and some parenting behaviors. Home-based programs had their greatest impact on parenting behaviors, child language, and parent participation in education and job training. The mixed-approach programs were most similar to the center-based programs, but also influenced children’s social and language development. EHS participation also seemed to have an energizing effect on participating families because they were more than twice as likely as control families to participate in parent education, parent-child, or parent support group activities, and they were more likely to benefit from key services available in their communities. It will be several years before the national evaluation of EHS reveals the full impact of EHS, but the preliminary analyses of the 24-month data are extremely encouraging. EHS appears to be having a positive impact both on child development and parental competence and selfsufficiency (Robinson & Fitzgerald, 2002).

Early Head Start and Home Visiting

The NICHD and EHS child care and child development programs each involved center-based and home-based components, reflecting the fact that exclusive center-based care during the early years—especially during infancy—is more the exception than the rule. For example, approximately 85% of infants receiving nonparental child care do so in their own homes, in the home of a care provider, or the home of a relative. In large measure, our knowledge of the effectiveness of early child care is generated from a database that is not representative of the population (an oversampling of low-income families), or of the contexts within which early child care takes place (an oversampling of center-based child care settings). Although there is great controversy in the literature about the effectiveness of home-visiting programs, the weight of that evidence is slowly shifting to questions of for whom and under what circumstances rather than whether home visiting is or is not effective in enhancing child development (Tableman, 2001). Moreover, as exemplified by two of the successful national demonstrations of child care for infants and toddlers that spanned low- to middle-class samples (EHS and NICHD), programs that combine some form of home visiting with other types of child care have generated some of the best benchmark guidelines for quality care to date.

However, there is an interesting paradox with respect to home visiting. Most home-visiting programs are designed to have a direct impact on parents and an indirect impact on the child. Of course, in many EHS programs involving home-based–center-based combinations or mixed models, there are both direct and indirect pathways leading to desired outcomes. In any of these cases, the home-based component is designed to enhance parenting skills and knowledge of child development, as well as the parent-infant relationship. Because home visitors work with parents to develop their parenting skills, the quality of their family relationships, and their ability to access human services within their communities, their immersion in child care can be as extensive as that of relatives or nonrelatives if one excludes the direct provision of care. Seldom have such programs been offered to care providers in nonparental child care settings even though infants and toddlers in such settings can have nonparental caregiving for as much as 40–50 hours per week (Smith, 2000). The concept of home visitor is implicit in the Head Start/Early Head Start training and technical assistance system designed to provide broad based support for all HS and EHS programs, regardless of the specific mix used by any specific program to deliver high-quality child care. Training and technical assistance staff visit program sites, assist with evaluating needs of program staff, and develop training opportunities related to identified needs (Mann, 2002). Training and technical-assistance on-site visits to programs during the implementation phase of EHS indicated that the time required to establish programs and to coordinate all of the necessary support components to assure quality care was underestimated. As a result, many programs needed to reorganize their approaches even while they were implementing new child care services (Mann, 2002). Because child care home-visitors help families to develop service delivery plans that are unique to family needs (parental age, parent-child interaction patterns, family resources, family composition) and that are extensions of the child care program, disruptions in the base child care program can have a negative impact on the home-visiting part of the child care service. For example, Gill, Greenberg, and Vazquez (2002) found that changes in program structure negatively affected home-visitor job satisfaction, work motivation, and staff turnover during the transitional period of program reorganization. Because adultchild relationships comprise a critical component of infanttoddler child care regardless of setting, studies of stability of care providers in child care centers reveal essentially the same findings as those found for home visitors (Raikes, 1993; Rubenstein, Pedersen, & Yarrow, 1977). Stability enhances the quality of the care-provider–child relationship and the social-emotional development of the child. Oppenheim, Sagi, and Lamb (1988) found that attachment security was related to personality characteristics of Israeli kindergarten children, but it did not predict specific characteristics such as egocontrol, empathy, achievement orientation, or independence. When changes in care provider do occur, infants and toddlers seem more likely to be negatively affected if change occurs during the second year of life (13–18 months) or if changes are frequent (Howes & Hamilton, 1993).

Although EHS programs are most likely to use infant mental health services or parents as teachers home visitor models, the Prenatal Early Infancy Project (PEIP) is one of the best-known home visitor intervention models because of the strong impact it obtained with a high-risk sample. The PEIP was developed to determine whether a home-visiting program could prevent poor developmental outcomes for infants of high-risk mothers (Olds 1988; Olds, Henderson, Chamberlin, & Tatelbaum, 1986; Olds, Henderson, Tatelbaum, & Chamberlin, 1986, 1988). Mothers were low-income, single, or teenage. Like the NICHD and EHS child care evaluations, PEIP families were randomly assigned to a treatment or comparison group. Families in the treatment group received home visits every 2 weeks during pregnancy and every 1–6 weeks for 2 years after birth. During the home visits, nurses provided parent education, helped mothers to develop informal support systems, and connected families with community services. The intervention was related to increased birth weight and gestation length of infants born to teen mothers and to smoking mothers. Throughout the first two postnatal years, mothers in the treatment group were significantly less likely to have verified child abuse than were mothers in the comparison group; they punished their children less, made fewer trips to the emergency room, and provided them with more appropriate toys (Olds, 1988). Mothers in the treatment group had greater increases in their knowledge of community services, use of social supports, and dietary improvements; they also attended more childbirth classes (Olds et al., 1986a, b) and returned to school sooner (Olds et al., 1988) than did mothers in the comparison group. One contributor to the success of PEIP was the relatively low turnover in home-visitor staff.

Carolina Abecedarian Project

The Carolina Abecedarian Project was designed to strengthen the intellectual competence and academic achievement of children from low-income families (Ramey & Campbell, 1984). For this project, 111 infants with single black mothers were randomly assigned either to an intervention group (57) or to a control group (54). Infants in the intervention group attended an educational child care program and the infants in the control group did not. All infants received infant formula, pediatric care, and supportive social services regardless of their assigned group. The preschool intervention provided early childhood education, pediatric care, and family support services beginning in infancy and lasting until the children entered kindergarten. The Abecedarian Project involved families with multiple risks (Ramey & Ramey, 1998), including poor maternal formal education, single-parent families, adolescent mothers, and authoritarian approaches to child rearing (Ramey, 2000; Ramey & Campbell, 1984, 1987; Ramey & Ramey, 1998.

The Abecedarian Project shared common characteristics with other early child care intervention programs, including programming that was connected to other local services for high-risk families, low child-to-teacher ratios, weekday programming available 10 hours per day (7:30–5:30), use of developmentally appropriate practices involving socialemotional and cognitive-language skills, and family support programs including pediatric care and nutritional supplementation. Because children in the control group also received health care assistance, social services, and nutritional supplements, any differences between groups can more readily be attributed to differences in the quality of early child care the children experienced (Ramey, 2000).

At 18, 24, 36, and 48 months of age, children in the intervention group scored higher on measures of cognitive functioning. By 4 years of age, children in the control group were six times more likely to score in the mild mental retardation range of the Stanford-Binet IQ test. When the infants reached preschool age, another randomization took place and children in the original intervention and control groups were assigned to new intervention and control groups. Follow-up studies conducted when the children were 12 indicated that intervention children did significantly better on measures of overall and Verbal IQ, but not Performance IQ. Children in the control group, however, were more likely to have IQ scores in the borderline range for mild mental retardation (F. A. Campbell & Ramey, 1990). Another follow-up performed when the children were 15 indicated that children in the preschool intervention group scored higher on measures of reading and mathematics than did children in the control group. Moreover, more children from the control groups were retained at grade level than were children from the intervention group (F. A. Campbell & Ramey, 1994). By 21 years of age, individuals in the treatment group were more likely to be in school (40% vs. 20%), and three times as many young adults from the treatment group (than from the control group) had either graduated from college or were still pursuing their degrees (35% vs. 12 %). Moreover, only 12% of the children in the treatment group received special education versus 48% of the control group.

Parent-Child Development Centers

The parent-child development centers (PCDCs), first established in 1969–1970, were designed to provide low-income families with support for child development as well as parent education. Mothers were the main targets for educational services, including information on child development, health, nutrition, parenting skills, and how to access community services and build social networks. Sites varied across many dimensions, but most notably in service delivery, ethnic group, program intensity, and home-based versus center-based contexts. A pre- and posttest evaluation strategy was used to evaluate program outcomes. At two of the sites, program participants scored higher than did controls on measures of positive maternal behavior. Dokecki, Hargrove, and Sandler (1983) reported differences in 36-month IQ scores between children in the program groups and those in control groups, with program children scoring significantly higher. PCDCs received negative evaluations for the quality of their facilities, lack of space, high child-staff ratios, low language interactions between parent and child, high dropout rates, and limited health services. Program strengths included positive affective environments, high-quality home-based services, high use of educational services, and organized support for the program. Initial program effects were positive but declined at kindergarten age. Despite these strengths, no differences were found later in IQ or school achievement as a function of the early preschool curriculum.

One of the Parent-Child Development Centers (PCDCs), the Houston Parent-Child Development Center, targeted primarily Mexican-American parents and their 12-month-old children (Johnson, 1988; Johnson & Breckenridge, 1982; Johnson & Walker, 1987, 1991). Emphasis was given to enhancing the positive affective relationships between mothers and their children. During the first year, families received home visits from trained neighborhood peer-educators. Weekend family workshops focused on issues of communication and support. When the children were 24 months of age, they attended a nursery school program while their mothers participated in classes on home management, personal development, and child development. After the 2-year intervention, PCDC mothers had more supportive interactions with the children than did comparison mothers. PCDC children scored better on the Stanford-Binet Intelligence Scale than did the comparison children. Follow-up studies were conducted when children were between 4 and 7 years of age and when they were 8–11 years of age. At the first followup, mothers of children in the comparison group reported higher levels of behavior problems—especially for boys— than did mothers of children in the PCDC group. At the second follow-up, teachers rated children in the comparison group—especially boys—as having more behavior problems than did children in the PCDC group.

Goteborg Child Care Study

The Goteborg study involved 140 Swedish 12- to 24-monthold toddlers who were divided into three groups (centerbased, family day care, and home cared). Fifty-two of the original sample were followed prospectively through 15 years of age (J. J. Campbell, Lamb, & Hwang, 2000). Over the course of the study, 43 children remained enrolled in family day care (13) or center day care (30), and 9 others changed from family day care to center day care. Data were collected on the quality of the home environment, parent demographics, quality of the supplemental child care setting, and target child peer interactions. Follow-up assessments were made 1 and 2 years after the initial data collection. Social competence was also assessed when the children were 6.5 and 8.5 years old. Final data collection occurred when the children were 15 years old.

Investigators found strong evidence of stability in child behavior over the span of 3–15 years of age as measured by indicators of social competence with peers during early development. However, at both 8.5- and 15-year assessments, neither the quality of home care nor the quality of supplemental child care was associated with social competence. There are striking parallels between stability of social competence in the Goteborg sample and stability of aggression in Dunedin, New Zealand children over a 20-year span (Caspi, Moffit, Newman, & Silva, 1996) and for externalizing behavior problems over a slightly shorter span for children in Michigan (Wong, Zucker, Puttler, & Fitzgerald, 1999).

Summary: Early Child Care and Child Development

A rich body of evidence has emerged that speaks to the positive effects of quality supplemental care on early child development. These research studies emphasize child development, parent involvement, and community networking. Some use random assignment evaluation designs and focus on different types of nonparental care—center-based, home based, and a combination of the two. Others use designs that allow for comparison of the effects of type of care on development. The most illustrative examples of this work include the NICHD Study of Early Child Care, theAdministration for Children,Youth, and Families national evaluation of Early Head Start; the Parent-Child Development Centers, the Carolina Abecedarian Project, and the Goteborg (Sweden) Child Care Study. Collectively, these studies provide information about the type of nonparental care children receive, the quality of that care, and the impact of nonparental care on child outcomes. Additionally, these studies shed light on the effect of comprehensive intervention on child outcomes over time. In general, the majority of infants start some type of nonparental care by the age of 4 months, experience multiple arrangements over the first year of life, and have working parents. More positive caregiving occurs when children are in smaller groups; child-to-adult ratios are lower; caregivers hold less authoritarian beliefs about child rearing; and physical environments are safe, clean, and stimulating. Poor children experience poor-quality family care but good-quality center care. Some studies report that center care has a modest positive effect on cognitive and language skills and reduction of aggression and other behavior problems. Moreover, family characteristics and the quality of mother-child relationships are stronger predictors of child outcomes than are child care factors; hence, interventions— including education and knowledge of child development— that have an effect on parent behaviors are critical in having a positive effect on children.

Factors Influencing Outcomes in Early Child Care Research

Regardless of whether supplemental infant care is delivered via home visiting, child care centers, or some mixed model, evaluation of early child care programs must take into account factors beyond those associated only with program components. Recall the conclusion reached by members of the National Research Council and Institute of Medicine (2000) that the question is not whether early preventive-intervention programs work, but for whom and under what circumstances. This clearly was recognized in the national evaluations of the NICHD Child Care Study and EHS, in which considerable effort was devoted to assessing characteristics of children and parents, parent-child interactions, and the ecological contexts within which development takes place. The list of such variables includes community characteristics (violence, community resources), family characteristics (poverty, single parenthood), parent characteristics (education level, substance abuse, depression or other psychopathology, employment), and child characteristics (disabilities, temperament). We believe that there are several critical areas in need of substantive research with respect to their implications for nonparental child care, regardless of whether that care is delivered in centers (including family child care), through home-based programs, or in mixed models. These areas involve child temperament and its impact on the caregiver-child relationship, mother-child social-emotional relationships, and father involvement in child and family development.

Temperament

Temperament refers to individual differences in emotional, motor, and attentional reactivity, as well as differences in selfregulation. The dominant view expressed in the literature is that such individual differences are biologically based, relatively stable over time and situations, and  evident fairly early in infancy (Strelau, 1998). References to variation in temperament date back to the earliest writings of Greek physicians.AlthoughAllport (1937) was among the first to contribute a formal definition of temperament, it was the work of Thomas and Chess (1977) that sparked contemporary research on the role that infant temperament may play in the development of parent-infant relationships. Instead of treating temperament as a characteristic anchored exclusively in biology or personality, Thomas and Chess conceptualized the impact of temperament within the framework of parent-infant interactions, and considered goodness of fit between the child’s temperament and the child’s parenting environment to be a major influence on the development of normative and atypical behavior (Chess & Thomas, 1986, 1991; Thomas, Chess, & Birch, 1968).

Thomas and Chess began to use the term temperament— originally referred to as individual reaction patterns—to describe individual differences in behavioral styles in relation to children’s interactions with their environments, particularly their caregiving environments. The impact of temperament resides in the goodness of fit between the child’s (perceived) temperament and the demands and expectations of the rearing environment. A good match is more likely to result in minimal conflict and stress between the child and significant adults in the caregiving environment, and a poor match is likely to exacerbate conflict and tension between the child and caregiver (Chess & Thomas, 1991). The bulk of the research has investigated the implications of temperament for mother-child interactions and relationships; little is known about temperament in relation to father or child care provider relationships.

The lack of knowledge about the impact of care-provider– infant goodness of fit is disconcerting when one considers that many infants and toddlers in center-based care spend as many as 50 hours each week in out-of-home settings. This may be especially troublesome for infants who are characterized as having difficult temperaments.

Thomas and Chess (1986) conceptualized difficult temperament as consisting of five components: irregularity, withdrawal from novelty, slow or no adaptation to change, negative mood, and intense emotional reactivity. Since their initial characterization, many investigators have defined difficult temperament differently, using a few as one trait and as many as 10 (Windle, 1991). The search for a finite set of operational components of difficult temperament is constrained by the social and perceptual qualities of temperament. Thus, what may be perceived to be difficult by one observer may be perceived to be positive by another. The care provider may be bothered by an assertive or somewhat uninhibited child, whereas the child’s father may perceive these traits as positive and in fact reward them. It is possible, however, that children with difficult temperaments actually display more stability across situations. For example, difficult temperament has been linked etiologically to behavioral undercontrol, especially in the context of high family dysfunction (Maziade, Caron, Cote, Boutin, & Thivierge, 1990); alcoholism and substance abuse (Mun, Fitzgerald, Puttler, Zucker, & von Eye, 2001); and poor academic functioning (Martin, 1989). Scarr and McCartney (1983) proposed a Patterson-type coercive model in which infant difficult temperament elicits negative parental responses, which in turn exacerbates the infant’s negative behaviors. Specifically, difficult temperament contributed to parental role dissatisfaction, insensitive caregiving, and an insecure infant-caregiver attachment relationship. Studying inner-city, low-income African American mothers and their infants, Ispa, Fine, and Thornberg (in press) found that infant difficult temperament and mother stress reaction were independently and inversely related to attachment security: high difficult temperament was related to low attachment security. Galinsky, Howes, Kontos, and Shinn (1994) found that more than 50 percent of the infants in family day care settings had insecure attachment relationships with their care providers, although this finding is not specifically linked to difficult temperament. Degree of difficulty is not the only aspect of temperament that may affect the quality of the care-provider–infant relationship. Fein (1995) reported that infants who were happy and socially detached received less attention from care providers, suggesting that such infants may fail to provide the normative cues necessary to elicit appropriate caregiver behavior.

Caregiver-Child Social-Emotional Relationships

No issue related to nonparental child care attracts more debate and heated discussion than does attachment. Attachment refers to the social-emotional relationship constructed by infants and their primary caregivers during the first year of life that functions to facilitate protection, survival, and reproductive fitness (Berlin & Cassidy, 2000).According to Bowlby (1969), the functional significance of attachment is expressed in proximity seeking (protection, care), avoidance of danger (fear, wariness), connectedness (social and interpersonal behavior), and novelty seeking (exploratory behavior). Attachment relationships during infancy are focused on relatively few individuals, with most infants showing a preference hierarchy when multiple caregivers are available. Typically, the relationship between mothering and the mother-infant attachment relationship is stronger than is the relationship between fathering and the father-infant attachment relationship (van Ijzendoorn & De Wolff, 1997). The set goal of the attachment system is to induce a sense of security and comfort in the infant that will eventually be internalized and expressed as a working model or internal mental representation of self and others (Berlin & Cassidy, 2000).

Because relationships are dynamic and transactional, infants and caregivers are constantly challenged to adjust their behavior as they seek to achieve a secure attachment relationship. Bowlby referred to this as a goal correcting process. Everyday observations of infants and toddlers in interaction with their parents, older siblings, and kin readily confirm that they form attachment relationships with many adults. However, as previously indicated, not all attachments are equal! Although infants generally have a preference for mother over father in situations that evoke attachment behaviors, they prefer father over a stranger when mother is not present (Cohen & Campos, 1974). Regardless of who the attachment object is, infants and adults either successfully negotiate a secure attachment relationship or they fall into one of three variants of insecure attachment: avoidant, ambivalent, or disorganized. Because attachment theory predicts caregiver specificity with respect to attachment relationships, several questions become paramount: Does nonparental child care interfere with the development of secure attachments between mother and infant? Do infants develop attachment relationships with their nonparental caregivers? If so, do nonparental attachments compensate for insecure mother-infant attachment relationships?

Research by Howes and her colleagues clearly illustrates that infants establish attachment relationships with their care providers (Howes & Hamilton, 1992a, 1992b; Howes & Smith, 1995; Howes, Rodning, Galluzzo, & Myers, 1988; Phillips, Howes, & Whitebook, 1992). Moreover, low teacher turnover enhances stability of the attachment relationship (Howes & Hamilton, 1992b). However, the ease of establishing a positive and secure care-provider–infant relationship may depend in part on the degree of preparation care providers have for teaching infants and toddlers. For example, Galinsky, Howes, and Kontos (1995) found that attachment security increased when family day care providers participated in a training program designed to enhance their caregiving skills.

The largest and most comprehensive longitudinal research findings on attachment in relation to child care effects are from the NICHD child care study noted earlier. At 6, 15, 24, and 36 months, 1,364 socially and racially diverse children were assessed after birth and followed to age 6. Positive child caregiving and language stimulation contributed meaningfully to early cognitive and language development. The higher the quality of provider-child interaction, the more positive were the mother-child interactions, and the more sensitive and involved were the mothers over the first 3 years (National Institute of Child Health and Human Development Early Child Care Research Network [NICHD], 1994, 1996, 1997, 2000).

The longer the time that infants and toddlers spent in group care, the fewer positive interactions they had with their mothers at 6 and 15 months of age, and the less affection they showed with their mothers at 2 and 3 years. Family income, mother’s vocabulary, home environment, and parental cognitive stimulation were more important than was child care quality in predicting cognitive and language advancements. Children in center care made larger gains than did those in family child care homes. Children from ethnic minority groups were more likely to be cared for in settings that offered fewer opportunities for messy play, reading books, and active explorations than were children from other groups. Children reared in economically disadvantaged homes were more likely to be insecurely attached to their mothers. When mothers strongly endorsed statements supporting the possible benefits of maternal employment for children’s development, their infants were more likely to be insecurely attached, and these mothers were also observed to be less sensitive and responsive. Their children were in poorer-quality care at earlier ages and for more hours per week.

Infant child care per se (observed quality of care, amount of care, age of entry, and frequency of care starts) did not appear to be a risk factor for insecure attachment. Maternal sensitivity however, was: Mothers who were least sensitive and responsive had more infants classified insecure avoidant; (16–19%) and fewer secure (53–56%) compared with the most sensitive mothers (9–11% insecure avoidant, 12–14% disorganized, 60–65% secure babies). The lowest proportion of secure attachment was noted when maternal sensitivity and child care quality were both low. For children with less sensitive mothers, attachment security proportions were higher if the children were in high-quality care than if they were in low-quality child care.

Effects of Separation

Separation from parents is a daily event for children of all age levels during early childhood. Putting baby down for a nap or going to the grocery store, to work, for dinner, or to a party are all events that separate parents from children. With the exception of parental employment, the effects of everyday separations on the child’s behavior have not been extensively studied. However, everyday separations probably have an overall beneficial influence on the young child, encouraging autonomy, independence, and the development of social competence. This is not to say that toddlers and preschoolers will always accept their parents’departure. In fact, as parents are about to leave, it is quite common for toddlers to protest and to try to stay with parents.

There are separations that can have profoundly disruptive influences on the young child’s development. These include the prolonged separations associated with death, divorce, desertion by one parent, or the loss of both parents—as in institutionalization or prolonged absences due to illness or to work. Until recently, the study of the effects of prolonged separation on the child’s development has focused on maternal separation involving the institutionalization of illegitimate children. This research raised the basic question of What are the consequences of prolonged maternal deprivation on the child’s development? To some extent, the hesitancy of American parents to rush into massive group child care for infants and toddlers can be traced to the belief that group care is detrimental to normal growth and development. This belief is rooted in studies of institutionally reared infants that pointed out the dangers of poor institutional care and deprivation from meaningful relationships with significant caregivers (Goldfarb, 1945; Skeels, 1936; Spitz, 1965).

Prolonged exposure to poor institutional care is associated with apathy, despair, and a pronounced deficit in social responsiveness—what Spitz referred to as hospitalism. Moreover, the effects of institutionalization are strongest if the infant is institutionalized during the period when attachment to significant caregivers normally occurs (Spitz, 1965). Rutter (1979) argues that it is not separation per se that causes development of affectionless pathology in children, but rather it is the failure to develop a secure attachment or emotional bond with primary caregivers in the first place that interferes with social competence and personality development.

Institutionalization represents the extreme form of separation from primary caregivers and because it is a rare event in the Western world, relative to the care settings in which most of the world’s infants and toddlers are reared, the results of institutionalization may translate poorly to other contexts. The fact is that most infants and toddlers experience frequent separations from their primary caregivers and we know precious little about how infants and their families negotiate these separations. For example, Tronick, Winn, and Morelli (1985) report that by the third week after birth, Efe infants of Zaire are cared for by an average of 3.6 nonmaternal individuals per hour, occupying 39% of daylight hours. By 18 weeks of age, Efe infants experience an average of 4.6 individuals per hour, which occupies 58% of daylight hours. Although when infants are fussy they are more likely to be passed to their mother for comforting, this in fact happens less than half of the time. Among the Loogoli of East Africa, mother’s caregiving responsibilities are influenced by such factors as household density. In large households, mothers are less likely to be their infants’ caregiver (Munroe & Munroe, 1971). Group size, therefore, may be an important factor for establishing secure infant-caregiver relationships. Infant-caregiver ratio is a standard aspect of all definitions of quality. Presumably, the lower the ratio, the better the care provider-infant relationship should be.

IntheUnitedStates,Suwalsky,Klein,Zaslow,Rabinovich, and Gist (1987) conducted one of the few studies of naturally occurring mother-infant separations. They tracked mothers of firstborn infants over the course of the first year of life, focusing on six types of separation—three during daytime (nonrecurrent separation for nonreimbursed activities; recurrent separations for employment; recurrent separations for nonreimbursed activities) and three overnight (maternal vacation, maternal employment, maternal illness/hospitalization). The range of separations across the 144 participating families was extraordinary. For example, the number of recurrent separations for employment ranged from 1 to 749; the number of nonrecurrent separations for nonreimbursed activities ranged from 2 to 545. Reasons for separations included work, shopping, education, religious activities, and leisure activities. Excluded are all the instances when infants and caregivers are separated within the household because it is nap time, dinner preparation time, time to work in the yard, or time to clean the house. Suwalsky et al. found that the most frequent location for supplemental care during employment-related separations was the home of the care provider (family child care, neighbor, relative). Fathers and relatives provided 31% of the infant’s care. One of the questions of interest to the investigators concerned the effect of changes in maternal employment on everyday separations. We find it interesting that when mothers’ employment hours decreased, the likelihood of another type of separation increased because mothers spent more time in nonreimbursed activities outside the home.

Maternal Employment

Some researchers report a slight increase in insecure attachments, aggression, and noncompliance in infants whose mothers were employed full-time during the infant’s first year of life (Barglow, Vaughn, & Molitor, 1987; Belsky & Eggebeen, 1991; Belsky & Steinberg, 1978). Others report no relationship between maternal work status and the quality of infants’attachment to their mothers. Lerner and Castellino (2001) suggest that the effects of maternal employment on infant and toddler development are unclear because few investigators adequately account for a wide range of confounding variables, such as the quality and amount of nonparental care, the age at which infants or toddlers enter into nonparental care, the stress load of work and family responsibilities, the degree to which mother receives parenting and housework assistance from other adults in the home, and a variety of characteristics of the infant. Clinical evidence concerning the effects of prolonged maternal deprivation on early development has existed for at least 50 years (Spitz & Wolf, 1946). Moreover, sufficient contemporary evidence exists to support the contention that toddlers’ interpersonal competencies flow from their initial relationships with their caregivers and other significant adults (Berlin & Cassidy, 2000). If the quality of the infant’s attachment relationship to significant caregivers is damaged, then one would predict difficulties in intrapersonal (self-concept) and interpersonal (self-other) relationships. Therefore, it is reasonable to question whether maternal separations associated with going to work are in any way causally related to negative child outcomes.

Working mothers do seem to provide role models for their children that are different from those provided by working mothers(Lerner&Castellino,2001).Forexample,thedaughters of working mothers tend to be more independent and more achievement-oriented than do the daughters of nonworking mothers. Both the sons and the daughters of working mothers tend to assume greater responsibility for household chores and to develop more positive attitudes toward maternal employment than do the children of nonworking mothers. In addition, the husbands of working wives tend to become more involved with routine household tasks and with caregiving than do husbands whose wives do not work.

One key determinant of the effects of maternal employment on her children’s behavior is the degree to which the mother is satisfied with her work and with the alternative child care arrangements available for her children. In fact, there is some indication that mothers who are satisfied with their personal and work life are more competent mothers than are mothers who either do not work or are not satisfied with their way of life (see Hoffman, 1974; Lerner & Castellino, 2002). Unfortunately, the effects of maternal employment on infant and toddler behavior have not received the research attention they deserve, especially in cultures in which women comprise a substantive portion of the labor force. For example, in the United States and Sweden, substantial numbers of infants and toddlers of working mothers spend 20 to more than 35 hours per week in paid child care.

Although working mothers with children under 2 years of age comprise the smallest proportion of working mothers with children under 5 years of age, the numbers are increasing. Not only are more women earning college degrees, but more low-income women are entering the workforce because of welfare-to-work requirements. In the former case, more highly educated women are more likely to be able to afford high-quality child care, whereas more poorly educated women are more dependent upon other family members for child care assistance.

Grandparents have a special role in child care in lowincome families (Casper, 1994), and they provide about 30% of the care for all children under 5 years when mothers are at work. Fathers provide about 18% of the care for all children under 5 (Smith, 2000), and low-income fathers are more than twice as likely as nonpoor fathers to be primary caregivers when mothers are at work (Casper, 1997). Nevertheless, in 1995 there were 9,342,000 children under 5 years of age in the United States who received nonparental and nonrelative care (14.8% in day care centers, 13.5% in nursery or preschool, 3.0% in Head Start, and 12.6% in family day care). The average hours in nonparental, nonrelative care ranged from 18.4 to 32.6 hours per week (Smith, 2000).

Honig and Park (1993) found that the longer children experienced out-of-home care, the more likely their teachers were to rate them as instrumentally more aggressive. Moore (1990) reported a link between the number of hours of infanttoddler out-of-home care and teacher reports of behavior problems in school. The relationships pertained to boys but not to girls. Specifically, spending more than 15 hours per week in out-of-home care when the child was between 18 and 21 months of age was associated with more teacher-reported behavior problems in school.

Infant-toddler development is influenced by forces emanating from a broad set of familial and social-cultural sources and is far from being solely an outcome of the mother-infant relationship (Brooks-Gunn, Duncan, & Aber, 1997; BrooksGunn, Leventhal, & Duncan, 2000). Historical and cultural factors must be considered with respect to their influence on family life and gender role if the boundaries that bind infants and toddlers to their cultural contexts are to be fully understood (Jordon, 1997). It is culture that in large measure defines the characteristics of family life and defines the nature of gender role—including the role of the father.

Fathers and Nonparental Child Care

Although knowledge of the father’s impact on child development lags far behind knowledge of the mother’s impact, there is sufficient literature to support a number of working hypotheses related to fathers and their young children (Fitzgerald & Montañez, 2000, 2001). The literature supporting such generalizations disproportionately involves data obtained from white, middle-class fathers. Data from a much more ethnically diverse and economically depressed sample of fathers whose children are participating in the national evaluation of EHS show some interesting parallels (Fitzgerald, Berlin, et al., 2000). The EHS study provided an opportunity to contrast responses from three types of fathers: Residential biological, nonresidential biological, and social fathers (nonbiologically related men who are identified by the mother as having a father role in the life of the child). Preliminary analyses suggest that all fathers in the EHS study were emotionally invested in their children, although the degree of investment was less for social fathers than for biological fathers.

Whether such demonstrative signs of investment and involvement reflect deep emotional commitment or are expressions of transitory engrossment (Greenberg & Morris, 1974) remains to be determined.

Thirty Years of Research

The research of the past three decades consistently has demonstrated that fathers’ lack of participation in routine caregiving of infants is more a matter of performance than of competence (Parke & Sawin, 1976; Parke, Power, & Gottman, 1979). In the United States the number of fathers who are primary care providers for their children is higher than ever before, and cross-culturally, fathers in dual-parent families are spending more time in routine child care than their fathers did (Burghes, Clarke, & Cronin, 1997; Horn, 2000). However, it is also the case cross-culturally that when fathers are in the presence of mothers and their infants, they tend to yield authority of child care to mothers (Mackey, 1996).

Fathers are invested in their infants and they establish attachment relationships with them. Demonstrating that father-infant attachments occur or that fathers seem to be invested in their children no longer seem to be critical questions. The more important questions and the questions of greater significance concern the circumstances that affect the degree of investment and the quality of the father-infant relationship. Nearly every theory of development assumes that fathers play a key role in the sex role identification of their children. Research attention has been given to fathers’ play interactions with infants and toddlers, with the conclusion that fathers are more actively involved in gender role socialization than are mothers, particularly with respect to their playful interactions with their sons and daughters (Fitzgerald, 1977; Power, 1981). Others report that fathers are less likely to differentially interact with their sons and daughters during infancy, but that by the toddler years they have a clear preference for rough-and-tumble play with their sons rather than with their daughters (Yogman, 1982). At least by 20 months of age, toddlers also respond differently to the play of their mothers and fathers. They are more likely to expect fathers than mothers to initiate play interactions, and by 30 months of age toddlers are more cooperative, involved, excited, and interested in play with their fathers than in play with their mothers (Clarke-Stewart, 1977). Fathers take a more direct role in regulating toddler compliance than mothers do and are particularly directive with demanding compliance from boys (Power, McGrath, Hughes, & Manire, 1994). Finally, fathers are also more likely than are mothers to encourage their children to be competitive, to be independent, and to take risks (Hewlett, 1992).

The most accurate summary statement that can be made at present regarding the effects of paternal deprivation on infants and toddlers is that we have a good sense of what outcomes are correlated with father absence, but that we are a long way from isolating the causal variables that explain the correlations. Cabrera, Tamis-LeMonda, Bradley, Hofferth, and Lamb (2000, p. 128) identified five ways in which father absence may have an impact on child development: (a) without a father there is no coparent, (b) economic loss frequently accompanies single motherhood, (c) social isolation and social disapproval of single or divorced mothers and children may lead to emotional distress and less adaptive functioning, (d) abandonment may cause psychological distress in children, and (e) conflict between parents may negatively influence children’s social-emotional development.

What do these findings suggest for infants and toddlers who receive supplemental child care? It may mean that many infants and toddlers spend considerable portions of the nonparental child care day in the presence of women but not in the presence of men. The number of males versus the number of women actively involved in infant and toddler child care is minuscule. Cabrera et al. (2000) note that little is known about how men learn to be fathers. Having been reared in a father-present family seems to be important. Investigators have found that men whose fathers were involved in raising them are more positively involved with their own children (Hofferth, 1999). Regardless of marital status or father presence or absence, it is clear that children reared in dysfunctional family systems or in families characterized by high parental risk loads are more likely to have problems themselves. These risk loads include the presence of intergenerational aggression and antisociality (Shears, Robinson, & Emde, in press). Men who reported high antisocial behavior during their youth had low assessments of themselves as fathers. Conversely, men who reported having positive relationships with their own fathers viewed themselves as good fathers. Some investigators have suggested that sociocultural factors create barriers for male involvement in early child care settings such as Early Head Start (Fitzgerald & Montañez, 2000). Families headed by low-income single mothers might especially benefit by having greater opportunities for men to be actively involved in child care and development. If young children in child care settings rarely interact with men, they may be denied social learning experiences that may be important for learning how to be a father (Cabrera et al., 2000).

Fathers and Family Risk Load

Infant mental health focuses on issues related to emotional development and within that context, the literature suggests that fathers may play a crucial role in helping children learn how to control their emotionality. There is no question that men like children and that fathers are both invested in and involved with their children (Mackey, 1996). Investment and involvement increase dramatically around the toddler years as children’s motor and language skills push them into more active, physical, and arousing play interactions, especially with their father or father figure. Outcomes of such activity include enhanced emotional regulation, compliance to rules and authority, and the internalization of rule structures.

Both father absence and father antisocial behavior are linked to behavioral dysregulation in children, especially boys. Antisocial fathers model aggression and provide the context for children to internalize aggressive scripts or schemas as action plans that are linked to specific contexts. This puts children at high risk for poor school performance, poor peer relationships, substance abuse, poor cognitive functioning, and the intergenerational transmission of dysfunctional interpersonal relationships (Carmichel Olsen, O’Connor, & Fitzgerald, 2001; Caspi, Henry, McGee, Moffitt, & Silva, 1995; Dobkin, Tremblay, & Sacchitelle, 1997; Fitzgerald, Puttler et al., 2000; Loukas, Fitzgerald, Zucker, & von Eye, 2001). Evidence suggests that father acceptance-rejection predicts child development outcomes better than does mother acceptance-rejection (Rohner, 1998). The negative effects of father absence or dysfunction are exacerbated by risky rearing environments, including risky neighborhoods (Zucker et al., 2000).

Large-scale child care programs such as Early Head Start that actively promote father involvement appear to be on the right course. Summers et al. (1999) used multiple ethnographic approaches to obtain qualitative responses from lowincome fathers of EHS children about their roles as fathers. Fathers identified providing financial support, just being there, caregiving, play, teaching, disciplining, providing love, and protection as key roles. Some evidence suggests that the new perception of fatherhood is one of father as equal coparent (Pleck & Pleck, 1997), and this theme was evident in many of the father interviews. It remains to be determined whether involvement in EHS programs will be sufficient to offset the gradual distancing that occurs between nonresidential fathers and their children during the time of youth and adolescence (Nord & Zill, 1996). Nevertheless, fathers seem to contribute disproportionately to family risk load, and programs that may help to induce a greater sense of responsibility for fatherhood are likely to assist in the reduction of that load.

Summary: Factors Influencing Outcomes in Early Child Care Research

Several factors influence the outcomes in early child care research. The first factor concerns child temperament and its impact on the caregiver-child relationship. Temperament refers to individual differences in emotional, motor, and attentional reactivity, as well as differences in self-regulation. The impact of temperament resides in the goodness of fit between the child’s (perceived) temperament and the demands and expectations of the rearing environment. Agood match is likely to result in minimal conflict and stress between the child and significant adults, and a poor match is likely to exacerbate conflict and tension. Much is known about maternal perception of infant temperament and its impact on motherinfant relationships. However, relatively less is known about how perceptions of the father and nonparental caregivers may affect their interactions with infants.

The second factor involves mother-child social-emotional relationships. Attachment refers to the social-emotional relationship constructed by infants and their primary caregivers. Research clearly shows that infants establish attachment relationships with their care providers. The ease of establishing a relationship depends in part on the degree of preparation care providers have for teaching infants and toddlers. In both the United States and other cultures, most infants and toddlers experience frequent separations from their primary caregivers. Yet very little is known about how these separations affect the development of the child. For instance, the effects of maternal employment on child development are still unclear because few investigators control for important confounding variables such as the quality and amount of nonparental care or the age at which children enter into nonparental care.

The third factor concerns father involvement in child and family development. Fathers are more actively involved in gender role socialization than are mothers. They take a more direct role in regulating toddler compliance than mothers do and are more likely than mothers are to encourage their children to be competitive, to be independent, and to take risks. Father absence may have a profound impact on child development. If young children in child care settings rarely interact with men, they may be denied social learning experiences that are important for learning how to be a father. Literature on infant mental health suggests that father absence and father antisocial behavior are linked to adverse outcomes in children, such as behavioral dysregulation, poor school performance, poor peer relationships, poor cognitive functioning, and substance abuse. Fathers seem to contribute disproportionately to family risk load. Programs that induce a greater sense of responsibility for fatherhood are likely to assist in the reduction of that load.

Nonparental Child Care: Cost, Quality, and Assurance

Cost and Quality

Cost is a primary concern when choosing child care. Based on the Children’s Defense Fund 2000 Yearbook, full-time child care costs may range from $4,000 to $10,000 per year per child. In one modest-sized midwestern city, family child care ranges from about $3,000 to $7,000 per year per child. Data from the sites involved with the national evaluation of EHS indicate that $9,646 per year is necessary to provide high-quality care to infants and toddlers in full-year child care. Clearly, nonparental child care is expensive. For some families, approximately one fourth of their income can be consumed by child care. When one considers that half of the families in the United States with young children earn less than $35,000 annually and that families with two fulltime working parents, each earning minimum wages, earn less than $22,000 combined, it is clear that access to nonparental child care depends upon the availability of government subsidies. Data from the 1990 National Child Care Survey (Hofferth, 1995) suggest that parental income does not affect the availability of child care. However, the same survey indicated that few programs accept subsidized children. In the final analysis, working poor and working-class families end up paying a greater proportion of their income for child care than do middle-class families. As a result, they often have to depend on kinship care or on informal, lowquality child care.

The pay for child care providers is limited by the income of the families served and is often less than what is considered to be an acceptable wage. The average yearly salary of a child care worker is less than $15,000. There are limited federal subsidies available to assist parents with the costs of licensed child care, but only 1 in 10 eligible children receive the financial support needed. The availability of quality child care is limited by lack of funding subsidies for child care workers and working parents. Making less than a livable wage greatly reduces the number of trained child care providers, the number of available child care slots, and the ability of parents to work, while it simultaneously increases staff turnover.

Benchmarking Quality

In the United States, child care tends to be viewed as a personal issue to be solved at the family level. The minority view is that child care availability and quality are issues of national concern that require extensive government support. According to the Children’s Defense Fund 2000 Yearbook, approximately 76% of children age birth to 5 years are placed in child care, and 60% of the children are infants. This suggests that quality child care is both a private and public issue; adequate spaces for child care enable working parents to have a placement for their children. Meeting quality standards assures that their children will thrive as well as—or perhaps better than—they might in home care (Phillips, 1992; Phillips, Howes, & Whitebook, 1992; Philips, Lande, & Goldberg, 1990; Phillips & Adams, 2001).

Theoretically, quality child care should be associated with children who are cognitively, linguistically, and socially prepared for preschool and kindergarten—children who are ready to learn. Because of their interactions with peers, they should be as socially skilled as are children who are reared at home with siblings. Initial studies of the effects of infant day care reported that infants reared in day care scored higher on measures of cognitive, linguistic, and social competence than did home-reared infants (Robinson & Robinson, 1971; Fowler, 1972; Lally, 1973; Honig & Brill, 1970; Keister, 1970). Nearly all studies from this era reported that infants in child care formed their strongest attachments to their mother (Caldwell, Wright, Honig, & Tannenbaum, 1970; Ragozin, 1980), thus allaying concerns of critiques that group day care would lead to emotional damage. Some investigators reported that toddlers reared in child care were less attentive to peers than were home-reared toddlers (Kagan, Kearsley, & Zelazo, 1975). Others found no differences between day care and home-reared children with respect to peer relationships. Rubenstein and Howes (1979) note that peers play an important role in child care by serving as models for one another. For example, in their study, infants’ positive affect and competence at toy play were higher when they were interacting with peers than when interacting with adults.

The 1970s was also a time when investigators focused on so-called natural experiments in infant-toddler child care that were provided by many countries in the Soviet Union and Western Europe. Child care programs in Hungary, East Germany, Czechoslovakia, Yugoslavia, and the Soviet Union were described in great detail (Brackbill, 1962; Bronfenbrenner, 1962; Meers & Marans, 1968; Wagner & Wagner, 1971). One lesson learned from these reports was that infant-toddler child care settings and curricula will reflect broader social-cultural values, and therefore might not generalize easily to other cultures.

Because many of these early studies assessed the impact on infants and toddlers enrolled in university demonstration day care centers or in formal state-supported centers, the extent to which findings generalized to the everyday context was compromised. Thus, the basic questions concerning the impact of nonparental child care continue to be raised. Does early nonparental child care have detrimental effects on children’s social-emotional and cognitive development? Is there an optimal group size for facilitating the development of competence? Are same-age peer groups more effective than multiple-age peer groups? Are the relationships among children different in day care than in home care, regardless of the nature of the peer group composition? The questions tend to be especially pointed at the group care of infants and toddlers, but they apply equally well to the broad range of caregiving contexts within which very young children are reared by nonparental caregivers.

Some investigators report that infants who experience nonparental center-based child care have poorer social-emotional development and social behavior skills as preschoolers (Belsky, 1986, 1988) and are more likely to score higher on measures of behavioral control than are home-reared children. Other investigators find that consistency in child care may be an important mediator of such outcomes. For example, Ketterlinus, Henderson, and Lamb (1992) found no differences between infants and toddlers reared in nonparental day care and home-reared comparison children. Stability of placement seemed to be a critical component of the lack of differences in outcome; infants or toddlers in the Ketterlinus et al. study experienced two sustained years in stable day care settings. In a retrospective study of maternal reports of their children’s behavior, Burchinal, Ramey, Reid, and Jaccard (1995) did not find evidence to support heightened levels of externalizing or internalizing behavior problems in infants reared in day care. Anderson’s (1989) study of Swedish 8-year-old children who experienced nonparental child care as infants found similar results from teacher ratings. In fact, teachers rated the children who had experienced day care as more persistent, more independent, less anxious, and more verbal than children who did not have prior day care experience. Of course, one critical variable in studies tracing the effects of infant day care on later developmental outcomes concerns the quality of the child care they experienced. Howes (1990) found that low-quality infant child care predicted poor peer relationships as kindergartners, whereas high-quality care did not. For example, infants who developed secure relationships with their teachers were gregarious and less aggressive in peer play as 4-year-olds, whereas those who were dependent and socially withdrawn were more aggressive in their play interactions (see also Pianta & Nimetz, 1991).

Publication of optimal standards for high-quality child care occurs frequently, filling newsletters of national child development associations, parenting magazines, state and national agency bulletins, newspaper “living today” sections, and countless web pages. Criteria for high-quality child care include a staff well-trained in child development; a staff that promotes warm, sensitive, and responsive interactions with infants and toddlers; low staff-child ratios and small classroom groups; stability of caregivers over time (Howes & Hamilton, 1992a, 1993); good nutrition and health practices; developmentally appropriate curriculum practices; good provider-parent relationships; and stimulating and safe environments.

Despite almost consensus agreement on these criteria for quality, there have been few systematic efforts to determine whether such factors as group size, child-staff ratios (Belsky, 1990; Hayes, Palmer & Zaslow, 1990), or provider training (Arnett, 1989) actually do define quality care that in turn has positive impacts on child development. Blau (1996) analyzed data from 1,309 teachers from 227 day care centers that participated in the National Child Care Staffing Study. Blau found that in the best-fitting models, conventional benchmarks such as group size, child-staff ratios, and staff training had nearly negligible impacts on quality of child care. Staff training and education had some positive but low-order effects on quality. Child care quality was defined by the Early Childhood Environment Rating Scale (ECERS) for preschool classrooms and the Infant-Toddler Environment Rating Scale (ITERS) for infant-toddler programs (Harms & Clifford, 1980, 1986). Factor analysis of the items in the ECERS and ITERS revealed two aggregate scales, which Blau defined as CARE (Appropriate Caregiving: adult-child interaction, supervision, discipline), and ACTIVITY (Appropriate Activity: materials, schedule, activities as a proxy measure of classroom environment). The Arnett Caregiver Interaction Scale, used to rate teacher’s interactions with children, generated three indexes (SENSITIVE, HARSH, DETACHED), which were combined with the ACTIVITY and CARE scales to assess quality of child care. We present this level of detail (a) to specify what Blau defined as quality care and (b) to indicate that none of the variables assessed were linked to child outcome data. Whereas Blau has made an important first step toward a systematic analysis of quality care, the next steps must involve child development outcomes because they represent the ultimate product of interest in the child care system.

Standards for high-quality child care almost always include reference to relationships between parents and care providers. Indeed, in national child care initiatives such as EHS, parents must be an integral component of the operating organization structure. Seldom, however, are parents involved in any systematic way in the evaluation of child care quality. A retrospective study of parents whose infants were enrolled in an infant-toddler child care program attempted to determine whether parent’s perceptions of their child’s proximal behavior were in any way linked to their attitudes about their child’s early child care experience (Ledesma, Fitzgerald, & McGreal, 1980). Parents of all infants who had been enrolled in the child care center over a 9-year period were mailed questionnaires designed to elicit their opinions about their children’s past child care experience and their current developmental status.

Nearly all parents reported feeling guilty when first enrolling their infants in the center; most often, this involved parents’ concerns about their attachment relationship with their child. We find it interesting that parents reported that they did develop strong attachment relationships with their infants and that their current relationships were also strong. They noted that their infants shared attachments with other caregivers but not to the detriment of the parent-infant relationship. They especially noted that enrollment of their infants in day care resulted in less stress on the family during a time when supplemental care was essential. Ledesma et al. (1980) note that some parents reported that they had to deal with public bias against having their infant in day care:

…and I did feel guilty. Our daughter was only six months old when I went back to school and we had to continuously reassure ourselves that we were doing the right thing. I must have called the center five times a day during her first few months there. Relatives didn’t make it any easier. My mother-in-law (as well as one of my professors) kept referring to “mothers who abandon their babies” whenever she could work it into a conversation. (pp. 47–48)

Other concerns frequently expressed involved the degree of parental involvement in center activities, caregiver-infant ratios, maintenance of physical health, and amount of physical space available in the center for play areas. An overwhelming number of parents reported that it was the quality of the center staff that surfaced as the most important factor contributing to parents’ perceptions of quality child care. The competent, caring, and knowledgeable senior caregivers (both registered nurses with considerable personal and professional experience with infants and toddlers) seemed to allay any concerns that parents had about physical space, health maintenance, and child development.

We seem to “know” the defining characteristics of child care quality; yet this knowledge faces many contradictions. We know that highly trained staff are essential to quality, yet relatively few states regulate training or have minimal standards for staff competency. We know what quality child care is, yet we continue to conduct research in order to identify the components of optimal child care. In some respects researchers seem to be intent on identifying the just-adequate environment that will do no harm, rather than focusing on the specific individual, familial, and systemic variables that may predict child outcomes independent of or in interaction with particular child care settings.

Although the defining features of high quality nonparental childcarehavebeenidentified,thesefeaturesareprogramand personnel specific—that is to say, child characteristics, family characteristics, and neighborhood characteristics are not typically part of the quality formula. We are a long way from fully understanding how intra-individual and familial variables interact with the child care setting to influence child outcomes, although we have clear pathways marked to guide such research (Anderson, 1989; Howes, 1988; Howes & Hamilton, 1992a, 1992b, 1993; Howes, Hamilton, & Matheson, 1994; Ketterlinus et al., 1992; Lucas, 2001; Pianta & Nimetz, 1991; Raikes, 1993; Ramey, 2000; Ramey & Ramey,1998;Rubinsteinetal.,1977).Doinfantswithdifficult temperaments score higher on measures of aggressive behavior regardless of their child care experiences? Does high-quality child care provide protective factors for children who are reared in low-quality home environments? Does the quality of the teacher-child relationship promote effective social interaction skills? Is high-quality early child care sufficient to facilitate resilience factors and coping skills amongchildrenrearedinhigh-riskfamilyorneighborhoodenvironments? One of the strongest conclusions Yoshikawa (1994,1995)reachedinhisanalysisoftheeffectsofchildcare was that poor quality is associated with poor outcomes and that high quality is associated with positive outcomes— cooperation with adults, the ability to imitate and sustain positive interactions with adults, and early competence in reading and math.

Assurance: Regulatory Policies and Nonparental Child Care

In the United States, child care centers are rapidly adopting curricula that flow from the principles and practices that constitute developmentally appropriate practices for early childhood education (Bredekamp, 1987a, 1987b).

The developmentally appropriate child-centered approach to early childhood education is not universal (Boocock, 1995). The Maori of New Zealand and residents of many Asian countries view early childhood education as a direct, downward extension of formal schooling. Low child-adult ratios valued in the United States apparently are less meaningful in France and Japan with respect to attainment of high-quality care or positive child development outcomes. Children who attend preschool in developing countries have better cognitive skills and overall school performance when compared with children in the same countries who do not attend preschool (Boocock, 1995). We find it interesting that countries that report positive outcomes for early childhood education tend to have national policies that support universal preschool and high-quality programs. The United States spends heavily to support a wide variety of prevention programs (EHS, Head Start, Healthy Start, Even Start, Early Start, Success by Six, Early Intervention for Infants and Toddlers) that target children in families whose incomes fit the federal definition of poverty or who for other reasons are at high risk. Currently, 31 states also fund programs that specifically target families with very young children. Only recently have efforts emerged to try to link existing programs into more integrated systems for children of low-income families. For example, performance standards for EHS require programs to build community networks among service providers to facilitate easy access to available services for EHS families.

States with higher levels of child care quality are more likely to report higher economies, safer neighborhoods, and better schools than are states with lower ratings. Young, Marsland, and Zigler (1997) analyzed the regulatory standards for center-based child care in the United States. Noting that a 1980 review reported that no state met federal standards of quality for group composition, staff training, and program of care, Young et al. sought to determine how much change has occurred during the 1980s. Data indicate that relatively little progress was made over a 10-year period. With the exception of the number of programs scoring in the good category, the majority of states were rated poor to very poor (or unregulated) on group composition and caregiver qualifications. Indeed, only Minnesota was rated as high as minimally acceptable with respect to caregiver qualifications. A more recent analysis of state regulations indicates that the major of states do not have training requirements for either family child care providers or for teachers in child care centers.

Perhaps policies adopted by the United States Department of Defense (DOD) for child care programs available to families in the military should serve as benchmarks for quality assurance (N. D. Campbell, Applebaum, Martinson, & Martin, 2000; N. D. Campbell, 2000; Lucas, 2001). The DOD provides military families with options of full-day care, part-day care, hourly care, occasional care, and long-term care, depending on need. One half of all children in military child care programs are in infant-toddler programs, which provide child care beginning as early as the sixth postnatal week in center programs and the fourth postnatal week in family programs. Regardless of the type of care needed, provision of care and related services is organized in a seamless system with one point of entry for each family. Monitoring for quality assurance requires annual certification of health
and safety regulations, developmental programming, child abuse prevention efforts, and staff training. Four unannounced inspections occur annually, one of which involves a multidisciplinary team (Lucas, 2001)! Clearly, the DOD takes seriously its commitment to provide high-quality child care to families serving in the military.

Summary: Nonparental Child Care

Although there continues to be national ambivalence regarding the use of nonparental child care for very young children, it is likely that current employment trends for women of childbearing age are not likely to reverse anytime soon. The simple fact is that many parents—even if they desire to stay at home with their very young children—may not be able to do so for economic reasons. Why should these parents be penalized for making choices that directly affect their ability to contribute to taking care of themselves and their families? More policy options are needed that provide support for families needing to access such care options.

It seems that we have come to recognize this reality as we examine the system established to support those enlisted in the armed forces. The benefits of such a system have been considerable—not only for children who participate in care regulated by the DOD, but also for families and for the government.When parents feel comfortable about the quality and safety of care, it likely has a positive impact on productivity. We need to carefully consider how all aspects of the military child care system (cost, quality, and assurance) can provide a model for what should also be in place in the private sector. Although it is certainly true that federal-, state-, and even foundation-funded efforts have been established, much of this work seems inadequate to sufficiently address the need that presently exists for nonparental child care.

Researchers have an important role to play in this process. As we consider the fact that many families will continue to rely upon nonparental care, we need a generation of research that goes beyond the question of whether child care influences development; more studies are needed that seek to understand the pathways through which these settings exert their influence and how providers can in turn improve their settings as a result of such research.

Entering The Twenty-First Century

The so-called new school-age child is now around 40 years old, and the realities of the workforce, equality for women, and changing roles for men suggest that the number of infants and toddlers living in supplemental care settings during some significant portion of their lives will continue to increase through the first decade of the twenty-first century. Although it may no longer be meaningful to ask where preventive intervention programs work (e.g, National Research Council and Institute of Medicine, 2000), considerable fine detail needs specification if we are to achieve the quality of child care that will facilitate optimal development of infants and young children. Failure to determine the boundaries of quality and to demand that all nonparental child care experiences fall within those boundaries could have serious neurobiological, psychological, and social consequences for infants and toddlers.

We have suggested several focal domains relevant to specifying the consequences of nonparental child care. The first domain focuses on characteristics of the infant and on the ecological context within which they organize. We suggest that dominant theories of the importance of mother-child relationships fail to adequately address the ecological context within which most parenting takes place, contrasting studies of separation experiences in the laboratory with studies of separations that take place through the normal course of everyday life. Moreover, we suggest that questions raised about the importance of mother-child relationships also be examined with respect to care-provider–infant relationships. The second domain examines the impact of fathers on infant and toddler development. Theory and research on fathers needs to move beyond simple imitation of the vast literature concerning maternal influences on child development. Although many theoretical concepts may prove to apply equally to mothers and fathers, the level of our current knowledge of the impact of fathers on child development will benefit from openness and creativity among developmental researchers. The third domain addresses issues of risk. The dominant questions in child care research have focused as much on a political agenda contrasting at-home rearing with out-ofhome rearing as they have on the impact of child care on mother-infant relationships.

Developmentalists need to move beyond these issues and examine the impact of nonparental care on infants who are at biological risk, familial risk, community risk, or any combination. Part of this agenda requires intensive study of the impact of cultural diversity in parenting and child-rearing beliefs, attitudes, and practices as they play out against the realities of needed placements in nonparental care settings (Garcia Coll, 1990). Although research on infants and families defined as ethnic minorities in the United States is increasing, our knowledge of variability in developmental pathways for infants of color is woefully inadequate (Fitzgerald et al., 1999). Finally, we suggested that exemplary models for highquality care provide a base for setting standards that researchers can use to assess the impact of the child care context on infant and toddler development. Finally, to paraphrase Michael Rutter’s observation in 1976, perhaps it is time to consider that the standards that society applies to the consequences of nonparental child care should also be applied to the consequences of parental child care.

Bibliography:

1. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Erlbaum.
2. Allport, G. W. (1937). Personality: A psychological interpretation. New York: Holt.
3. Anderson, B.-E. (1989). Effects of public day care: A longitudinal study. Child Development, 60, 857–866.
4. Arnett, J. (1989). Caregivers in day care centers: Does training matter? Journal of Applied Developmental Psychology, 10, 541–552.
5. Barglow, P., Vaughn, B. E., & Molitor, N. (1987). Effects of maternal absence due to employment on the quality of infant-mother attachment in a low-risk sample. Child Development, 53, 53–61.
6. Belsky, J. (1986). Infant day care: A cause for concern? Zero to Three, 6, 1–9.
7. Belsky, J. (1988). The effects of infant day care reconsidered. Early Childhood Research Quarterly, 3, 235–272.
8. Belsky, J. (1990). Parental and nonparental child care and children’s socioemotional development: A decade in review. Journal of Marriage and the Family, 52, 885–903.
9. Belsky, J., & Eggebeen, D. (1991). Early and extensive maternal employment and young children’s socioemotional development: Children of the National Longitudinal Survey of Youth. Journal of Marriage and the Family, 53, 1083–1098.
10. Belsky, J., & Steinberg, L. D. (1978). The effects of daycare: A critical review. Child Development, 49, 929–949.
11. Berlin, L. J., & Cassidy, J. (2000). Understanding parenting: Contributions of attachment theory and research. In J. D. Osofsky & H. E. Fitzgerald (Eds.), WAIMH handbook of infant mental health (pp. 131–170). New York: Wiley.
12. Bertalanffy, L. von (1950). An outline of General Systems Theory. British Journal of Philosophy and Science, 1, 139–164.
13. Bertalanffy, L. von (1968). General systems theory. New York: Braziller.
14. Blau, D. M. (1996). The production of quality in child care centers. The Journal of Human Resources, 32, 334–387.
15. Boocock, S. S. (1995). Early childhood programs in other nations: Goals and outcomes. The Future of Children, 5, 94–114.
16. Bowlby, J. (1969). Attachment. New York: Basic Books.
17. Brackbill, Y. (1962). Research and clinical work with children. In R. A. Bauer (Ed.), Some views on Soviet psychology (pp. 99– 164). Washington, DC: American Psychological Association.
18. Bredekamp, S. (Ed.). (1987a). Accreditation criteria and procedures of the National Academy of Early Childhood Programs. Washington, DC: National Association for the Education of Young Children.
19. Bredekamp, S. (1987b). Developmentally appropriate practice in early childhood programs serving children from birth through age 8. Washington, DC: National Association for the Education of Young Children.
20. Bronfenbrenner, U. (1962). Soviet methods of character education: some implications for research. American Psychologist, 17, 550–564.
21. Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.
22. Bronfenbrenner, U., & Morris, P. (1998). The ecology of developmental processes. In W. Damon (Series Ed.) and R. M. Lerner (Vol. Ed.), Handbook of child psychology: Vol. 5. Theoretical models of human development (5th ed., pp. 993–1028). New York: Wiley.
23. Brooks-Gunn, J., Duncan, G. J., & Aber, J. L. (1997). Neighborhood poverty (Vol. 1). New York: Russell Sage.
24. Brooks-Gunn, J., Leventhal, T., & Duncan, G. J. (2000). Why poverty matters for young children: Implications for policy. In J. D. Osofsky & H. E. Fitzgerald (Eds). WAIMH handbook of infant mental health: Vol. 3. Parenting and child care (pp. 89– 132). New York: Wiley.
25. Burghes, L., Clarke, L., & Cronin, N. (1997). Fathers and fatherhood in Britain. London: Family Policy Studies Centre.
26. Burchinal, M. R., Ramey, S. L., Reid, M. K., & Jaccard, J. (1995). Early child care experiences and their association with family and child characteristics during middle childhood. Early Childhood Research Quarterly, 10, 33–61.
27. Cabrera, N. J., Tamis-LeMonda, C. S., Bradley, R. H., Hofferth, S., & Lamb, M. E. (2000). Fatherhood in the twenty-first century. Child Development, 71, 127–136.
28. Caldwell, B. M., Wright, C. M., Honig, A. S., & Tannenbam, J. (1970). Infant day care and attachment. American Journal of Orthopsychiatry, 40, 397–412.
29. Campbell, F. A., & Ramey, C. T. (1990). The relationship between Piagetian cognitive development, mental test performance, and academic achievement in high-risk students with and without early educational experience. Intelligence, 14, 293–308.
30. Campbell, F. A., & Ramey, C. T. (1994). Effects of early intervention on intellectual and academic achievement: A follow-up study of children from low-income families. Child Development, 65, 684–698.
31. Campbell,J.J.,Lamb,M.E.,&Hwang,C.P.(2000).Earlychild-care experiences and children’s social competence between ¹₂ and 15 years of age. Applied Developmental Science, 4, 166–175.
32. Campbell, N. D. (2000, May). Military provides model for child care reforms. Department of Defense News Briefing. Washington, DC: U.S. Department of Defense.
33. Campbell, N. D., Applebaum, J. C., Martinson, K., & Martin, E. (2000, April). Be all that we can be: Lessons from the military for improving our nation’s child care system. Washington, DC: National Women’s Law Center.
34. Carlson, N. A., & Cassell, T. Z. (1984). Asocio-ecological model of adaptive behavior and functioning. In H. E. Fitzgerald, B. M. Lester, & M. W. Yogman (Eds.). Theory and research in behavior pediatrics (Vol. 2, pp. 31–66). New York: Plenum Press.
35. Carmichael Olson, H., O’Connor, M. J., & Fitzgerald, H. E. (2001). Lessons learned from study of the developmental impact of parental alcoholism. Infant Mental Health Journal, 22, 271–290.
36. Casper, L. M. (1994). Who’s minding our preschoolers? Fall 1994 update. Current Population Reports. Washington, DC: U.S. Department of Commerce.
37. Casper, L. M. (1997). My daddy takes care of me! Fathers as care providers. Current Population Reports. Washington, DC: U.S. Department of Commerce.
38. Caspi, A., Henry, B., McGee, R. O., Moffitt, T. E., & Silva, P. A. (1995). Temperamental origins of child and adolescent behavior problems: From age three to age fifteen. Child Development, 66, 55–68.
39. Caspi, A., Moffitt, T. E., Newman, D. L., & Silva, P. A. (1996). Behavioral observations at age 3 years predict adult psychiatric disorders. Archives of General Psychiatry, 53, 1034–1039.
40. Chess, S., & Thomas, A. (1986). Temperament in clinical practice. New York: Guilford Press.
41. Chess, S., & Thomas, A. (1991). Temperament and the concept of goodness of fit. In J. Strelau & A. Angleitner (Eds.). Explorations in temperament: International perspectives on theory and measurement (pp. 15–28). New York: Plenum Press.
42. Clarke-Stewart, K. A. (1977). Child care in the family: A review of research and some propositions for policy. New York: Academic Press.
43. Clarke-Stewart, K. A., Allhusen, V. D., & Clemens, D. C. (1995). Nonparental caregiving. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 3. Status and social conditions of parenting (pp. 151–176). Mahwah, NJ: Erlbaum.
44. Cochran, M. (Ed.). (1993). International handbook of child care policies and programs. Westport, CT: Greenwood Press.
45. Cohen, L. J., & Campos, J. J. (1974). Father, mother, and stranger as elicitors of attachment behaviors in infancy. Developmental Psychology, 10, 146–154.
46. Coley, R. L. (2001). Emerging research on low-income, unmarried, and minority fathers. American Psychologist, 56, 743–753.
47. Commissioner’s Office of Research and Evaluation, Head Start Bureau, Administration on Children, Youth, and Families, & Department of Health and Human Services (1999a, December). Leading the way: Characteristics and early experiences of selected Early Head Start Programs: Vol. 1. Cross-site perspectives. Washington, DC: U.S. Department of Health and Human Services.
48. Commissioner’s Office of Research and Evaluation, Head Start Bureau, Administration on Children, Youth, and Families, & Department of Health and Human Services (1999b, December). Leading the way: Characteristics and early experiences of selected Early Head Start Programs: Vol. 2. Program profiles. Washington, DC: U.S. Department of Health and Human
49. Commissioner’s Office of Research and Evaluation, Head Start Bureau, Administration on Children, Youth, and Families, & Department of Health and Human Services (2001a, December– June). Building their futures: How Early Head Start programs are enhancing the lives of infants and toddlers in low-income families: Vol. 1. Technical report. Washington, DC: U.S. Department of Health and Human Services.
50. Commissioner’s Office of Research and Evaluation, Head Start Bureau, Administration on Children, Youth, and Families, & Department of Health and Human Services (2001b, December– June). Building their futures: How Early Head Start programs are enhancing the lives of infants and toddlers in low-income families: Vol. 2. Technical report appendices. Washington, DC: U.S. Department of Health and Human Services.
51. Commissioner’s Office of Research and Evaluation, Head Start Bureau, Administration on Children, Youth, and Families, & Department of Health and Human Services (2001c, December– January). Building their futures: How Early Head Start programs are enhancing the lives of infants and toddlers in lowincome families: Summary report. Washington, DC: U.S. Department of Health and Human Services.
52. Dobkin, P. L., Tremblay, R. E., & Sacchitelle, C. (1997). Predicting boys’ early-onset substance abuse from father’s alcoholism, son’s disruptiveness, and mother’s parenting behavior. Journal of Consulting and Clinical Psychology, 65, 86–92.
53. Dokecki, P. R., Hargrove, E. C., & Sandler, H. M. (1983). An overview of the Parent Child Development Center social experiment. In R. Haskins & D. Adams (Eds.), Parent education and public policy. Norwood, NJ: Ablex.
54. Duncan, G. J., & Brooks-Gunn, J. (Eds.). (1997). Consequences of growing up poor. New York: Russell Sage.
55. Fein, G. G. (1995). Infants in group care: Patterns of despair and detachment. Early Childhood Research Quarterly, 10, 261–275.
56. Fenichel, E., & Mann, T. L. (2001). Early Head Start for lowincome families with infants and toddlers. The Future of Children, 11, 135–141.
57. Fitzgerald, H. E. (1977). Infants and caregivers: Sex differences as determinants of socialization. In E. Donelson & J. Gullahorn (Eds.), Women: A psychological perspective. New York: Wiley.
58. Fitzgerald, H. E., Berlin, L., Cabrara, N., Coker, D., Pan, B., Raikes, H., Roggman, L., Spellman, M., Tamis-LeMonda, C., & Tarrullo, L. (2000, July). Twenty-four months of fatherhood: Low-income men and their toddlers. Factors affecting paternal involvement in infant and toddler development: insights from early head start. In symposium (L. Roggman and K. Boller, chairs), 24 months of fatherhood: Low income men and their toddlers. Presented at the biennial meeting of the International Conference on Infant Studies, Brighton, UK.
59. Fitzgerald, H. E., Johnson, R. B., Van Eqeren, L. A., Castellino, D. R., Johnson, C. B., & Judge-hawton, M. (1999). Infancy and culture: An international review and source book. New York: Falmer Press.
60. Fitzgerald, H. E., & Montañez, M. (2000, October). Infant mental health and Early Head Start: Building capacity for father engagement. Paper presented at the Head Start Infant Mental Health Forum, “Addressing mental health needs of infants, parents, and families in Early Head Start and Migrant Head Start: Lessons from the scientific community,” Washington, DC.
61. Fitzgerald, H. E., & Montañez, M. (2001). Fathers as facilitators of infant mental health: Implications for Early Head Start. Zero to Three, 22, 25–28.
62. Fitzgerald, H. E., Puttler, L. I., Mun, E.-Y., & Zucker, R. A. (2000). Prenatal and postnatal exposure to parental alcohol use and abuse. In J. D. Osofsky & H. E. Fitzgerald (Eds.), WAIMH handbook of infant mental health: Vol. 4. Infant mental health in groups at high risk (pp. 124–159). New York: Wiley.
63. Fitzgerald, H. E., Strommen, E. A., & McKinney, J. P. (1977). The infant and young child. Homewood, IL: Dorsey Press.
64. Fitzgerald, H. E., Zucker, R. A., & Yang, H.-Y. (1995). Developmental systems theory and alcoholism: Analyzing patterns of variation in high risk families. Psychology of Addictive Behaviors, 11, 49–58.
65. Ford, D. H., & Lerner, R. M. (1992). Developmental systems theory: An integrative approach. Newbury Park, CA: Sage.
66. Fowler, W. (1972). A developmental learning approach to infant care in a group setting. Merrill-Palmer Quarterly, 18, 145–175.
67. Freud,S.(1946).Collectedpapers(J.Strachey,Trans.)(Vols.1,2,& 3). London: Hogarth Press.
68. Galinsky, E., Howes, C., & Kontos, S. (1995). The family child-care training study: Highlights of findings. New York: Families and Work Institute.
69. Galinsky, E., Howes, C., Kontos, S., & Shinn, M. (1994). The study of children in family child care and relative care. New York: Families and Work Institute.
70. Gill, S., Greenberg, M. T., & Vazquez,A. (2002). Changes in the servicedeliverymodelandhomevisitors’jobsatisfactioninanEarly Head Start. Infant Mental Health Journal, 23, 182–196.
71. Goldfarb, W. (1945). Effects of psychological privation in infancy and subsequent development. American Journal of Psychiatry, 102, 18–33.
72. Gottlieb, G. (1991). Experiential canalization of behavioral development: Theory. Developmental Psychology, 27, 4–13.
73. Greenberg, M., & Morris, N. (1974). Engrossment: The newborn’s impact upon the father. American Journal of Orthopsychiatry, 44, 520–531.
74. Harms, T., & Clifford, R. (1980). Early Childhood Environment Rating Scale. New York: Teachers College Press.
75. Harms, T., & Clifford, R. (1986). Infant-Toddler Environment Rating Scale. Chapel Hill: Frank Porter Graham Center.
76. Hayes, C. D., Palmer, J. L., & Zaslow, M. J. (Eds.). (1990). Who cares for America’s children? Child care policy for the 1990s. Washington DC: National Academy Press.
77. Hernandez, D. (1997, June). Findings about the changing face of childhood: family, economics, and ethnicity. Paper presented at the Science Writers Workshop, NIH Office of Behavioral and Social Sciences. Washington, DC: American Sociological Association.
78. Hewlett, B. S. (1992). Husband-wife reciprocity and the fatherinfant relationship among AKA pygmies. In B. S. Hewlett (Ed.), Father-child relations: Cultural and biosocial contexts (pp. 153–176). New York: de Gruyter.
79. Hofferth, S. L. (1995). Caring for children at the poverty line. Children and Youth Services Review, 17, 61–90.
80. Hoffereth, S. L. (1999). Race/ethnic differences in father involvement with young children: A conceptual framework and empirical test in two-parent families. Paper presented at the Urban Seminar on Fatherhood, Harvard University, Cambridge, MA.
81. Hoffman, J. M. (1994). The fish is in the water and the water is in the fish. The Signal, 2, 5–6.
82. Hoffman, L. W. (1974). Effects of maternal employment on the child: A review of research. Developmental Psychology, 10, 204–228.
83. Honig, A. S. (1995). Choosing child care for young children. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 4. Applied and practical parenting (pp. 411–436). Mahwah, NJ: Erlbaum.
84. Honig, A. S., & Brill, S. A. (1970, September). A comparative analysis of the Piagetian development of twelve-year-old disadvantaged infants in an enrichment center with others not in such a center. Paper presented at the meeting of the American Psychological Association, Miami, FL.
85. Honig, A. S., & Park, K. (1993). Effects of day care on preschool sex-role development. American Journal of Orthopsychiatry, 63, 481–486.
86. Horn, W. F. (2000). Fathering infants. In J. D. Osofsky & H. E. Fitzgerald (Eds.), WAIMH handbook of infant mental health: Vol. 3. Parenting and child care (pp. 269–298). NewYork:Wiley.
87. Hornby, H., Zeller, D., & Karraker, D. (1996). Kinship care in America: What outcomes should policy seek? Child Welfare, LXXV, 397–418.
88. Howes, C. (1988). The peer interactions of young children. Monographs of the Society for Research in Child Development, 53(1, Serial No. 217).
89. Howes, C. (1990). Can the age of entry and the quality of infant child care predict adjustment in Kindergarten? Development Psychology, 26, 252–303.
90. Howes, C., & Hamilton, C. E. (1992a). Children’s relationships with caregivers: Mothers and child care teachers. Child Development, 63, 859–866.
91. Howes, C., & Hamilton, C. E. (1992b). Children’s relationships with child care teachers: Stability and concordance with parental attachments. Child Development, 63, 867–878.
92. Howes, C., & Hamilton, C. E. (1993). The changing experience of child care: Changes in teachers and in teacher-child relationships and children’s social competence with peers. Early Childhood Research Quarterly, 8, 15–32.
93. Howes, C., Hamilton, C. E., & Matheson, C. C. (1994). Children’s relationships with peers: Differential associations with aspects of the teacher-child relationship. Child Development, 65, 253– 263.
94. Howes, C., Rodning, C., Galluzzo, D. C., & Myers, L. (1988). Attachment and child care: Relationships with mother and caregiver. Early Childhood Research Quarterly, 3, 403–416.
95. Howes,C.,&Smith,E.W.(1995).Childrenandtheirchildcarecaregivers: Profiles of relationships. Social Development, 4, 44–61.
96. Ispa, J. M., Fine, M. A., & Thornberg, K. R. (2002). Maternal personality as a moderator of relations between difficult infant temperament and attachment security in low-income families. Infant Mental Health Journal, 23, 130–144.
97. Jerald, J. (2000). Early Head Start. National Head Start Bulletin, 69, 1–3.
98. Johnson, D. J. (1988). Parental racial socialization strategies of Black parents in three private schools. In D. T. Slaughter & D. J. Johnson (Eds.). Visible now: Blacks in private schools (pp. 251– 267). Westport, CT: Greenwood.
99. Johnson, D. L., & Breckenridge, J. N. (1982). The Houston Parentchild Development Center and the primary prevention of behavior problems in young children. American Journal of Community Psychology, 10, 305–316.
100. Johnson, D. L., & Walker, T. (1987). Primary prevention of behavior problems in Mexican-American children. American Journal of Community Psychology, 15, 375–385.
101. Johnson, D. L., & Walker, T. (1991). A follow-up evaluation of the Houston Parent-Child Development Center: School performance. Journal of Early Intervention, 15, 226–236.
102. Jordan, B. (1997). Gender, politics and infant mental health. The Signal, 3, 12–13.
103. Kagan, J., Kearsley, R. B., & Zelazo, P. R. (1975). The emergence of initial apprehension to unfamiliar peers. In M. Lewis & L. Rosenblum (Eds.), Friendship and peer relations (pp. 187– 206). New York: Wiley.
104. Kamerman, S. B., & Kahn, A. J. (1995). Innovations in toddler day care and family support services: An international overview. Child Welfare, LXXIV, 1281–1300.
105. Keister, M. E. (1970). A demonstration project: Group care of infants and toddlers. Final report submitted to the Children’s Bureau, Office of Child Development. Washington, DC: U.S. Department of Health, Education, and Welfare.
106. Ketterlinus, R. D., Henderson, S. H., & Lamb, M. E. (1992). Les effets du type de garde de l‘emploi maternel et del’estime de soi sur le comportement des enfants [The effect of type of child care and maternal employment on children’s behavioral adjustment and self-esteem]. In B. Pierrrehumbert (Ed.), L’accueil du jeune enfant: Politiques et recherches dans les differents pays [Child care in infancy: Policy and research issues in different countries] (pp. 150–163). Paris: Les Editions Sociales.
107. Lally, R. J. (1973). The family development research program: A program for prenatal, infant, and early childhood enrichment. College for Human Development. Syracuse University, Syracuse, NY.
108. Lamb, M. E. (1998). Nonparental child care: Context, quality, correlates, and consequences. In I. E. Sigel & K. A. Renninger (Eds.), Handbook of child psychology: Vol. 4. Child psychology in practice (pp. 73–133). New York: Wiley.
109. Ledesma, S., Fitzgerald, H. E., & McGreal, C. E. (1980). Parent’s perceptions of the infant’s day care experience. Infant Mental Health Journal, 1, 42–55.
110. Lerner, J. V., & Castellino, D. R. (2002). Daycare and maternal employment in the 21st century: Conflicts and consequences for infant development. In H. E. Fitzgerald, K. Karraker, & T. Luster (Eds.), Infant development: Ecological perspectives (pp. 143– 164). New York: Routledge.
111. Loukas, A., Fitzgerald, H. E., Zucker, R. A., & von Eye, A. (2001). Parental alcoholism and co-occurring antisocial behavior: Prospective relationships to externalizing behavior problems in their young sons. Journal of Abnormal Child Psychology, 29, 91–106.
112. Lucas, M.-A. (2001). The military child care connection. The Future of Children, 11, 129–133.
113. Mackey, W. C. (1996). The American father: Biocultural and developmental aspects. New York: Plenum Press.
114. Mann, T. L. (2002). Training and technical assistance for Early Head Start. Infant Mental Health Journal, 23, 36–47.
115. Martin, R. P. (1989). Activity level, distractibility and persistence: Critical characteristics in early schooling. In G. A. Kohnstamm, J. E. Bates, & M. K. Rothbart (Eds.), Temperament in childhood (pp. 451–461). Chichester, UK: Wiley.
116. Maziade, M., Caron, C., Cote, R., Boutin, P., & Thivierge, J. (1990). Extreme temperament and diagnosis. Archives of General Psychiatry, 47, 447–484.
117. Meers, D. R., & Marans, A. E. (1968). Group care of infants in other countries. In L. L. Dittmann (Ed.), Early child care (pp. 237– 282). New York: Atherton Press.
118. Miller, J. G. (1978). Living systems. New York: McGraw-Hill.
119. Morelli, G. A., & Verhoef, H. (1999). Who should raise my child? A cultural approach to understanding nonmaternal child care decisions. In L. Balter & C. S. Tamis-LeMonda (Eds.), Child psychology: A handbook of contemporary issues (pp. 491–509). New York: Taylor and Francis.
120. Mun, E-Y., Fitzgerald, H. E., Puttler, L. I., Zucker, R. A., & von Eye, A. (2001). Temperamental characteristics as predictors of externalizing and internalizing child behavior problems in the contexts of high and low parental psychopathology. Infant Mental Health Journal, 22, 393–415.
121. Munroe, R. H., & Munroe, R. L. (1971). Household density and infant care in an East African society. Journal of Social Psychology, 83, 3–13.
122. National Research Council and Institute of Medicine (2000). From Neurons to neighborhoods: The science of early childhood development (J. P. Shonkoff & D. A. Phillips, Eds.). Washington, DC: Board on Children, Youth, and Families, Commission on Behavioral and Social Sciences and Education.
123. National Institute of Child Health and Human Development Early Child Care Research Network. (1994). Child care and child development: The NICHD study of early child care. In S. L. Friedman & H. C. Haywood (Eds.), Developmental follow-up: Concepts, domains and methods (pp. 377–396). New York: Academic Press.
124. National Institute of Child Health and Human Development Early Child Care Research Network. (1996). Characteristics of infant child care: Factors contributing to positive caregiving. Early Childhood Research Quarterly, 11, 269–306.
125. National Institute of Child Health and Human Development Early Child Care Research Network. (1997). Familial factors associated with the characteristics of nonmaternal care of infants. Journal of Marriage and the Family, 59, 389–408.
126. NICHD Early Child Care Research Network. (2000). Characteristics and quality of child care for toddlers and preschoolers. Applied Developmental Science, 4, 116–135.
127. Nord, C. W., & Zill, N. (1996). Non-custodial parents’participation in their children’s lives: Evidence from the Survey of Income and Program Participation (2 vols.). Final report prepared for the office of the Assistant Secretary of Planning and Evaluation. Washington, DC: U.S. Department of Health and Human
128. Olds, D. L. (1988). The prenatal/early infancy project. In R. H. Price (Ed.), Fourteen ounces of prevention (pp. 9–23). Washington, DC: American Psychological Association.
129. Olds, D. L., Henderson, C. R., Chamberlin, R., & Tatelbaum, R. (1986a). Preventing a child abuse and neglect: A randomized trial of nurse home visitation. Pediatrics, 78, 65–78.
130. Olds, D. L., Henderson, C. R., Tatelbaum, R., & Chamberlin, R. (1986b). Improving the delivery of prenatal care and outcomes of pregnancy: A randomized trial of nurse home visitation. Pediatrics, 77, 16–28.
131. Olds, D. L., Henderson, C. R., Tatelbaum, R., & Chamberlin, R. (1988). Improving the life-course development of socially disadvantaged mothers: A randomized trial of nurse home visitation. American Journal of Public Health, 78, 1436–1444.
132. Oppenheim, D., Sagi, A., & Lamb, M. E. (1988). Infant-adult attachments on the kibbutz and their relation to socioemotional development four years later. Developmental Psychology, 24, 427–433.
133. Osofsky, J. D., & Fenichel, E. (Eds.). (1994). Caring for infants and toddlers in violent environments. Arlington, VA: Zero to Three/National Center for Clinical Infant Programs.
134. Parke, R. D., Power, T. G., & Gottman, J. M. (1979). Conceptualization and quantifying influence patterns in the family triad. In M. E. Lamb, S. J. Suomi, & G. R. Stephenson (Eds.), Social interaction analysis: Methodological issues (pp. 231–253). Madison: University of Wisconsin Press.
135. Parke, R. D., & Sawin, D. B. (1976). The father’s role in infancy: A reevaluation. The Family Coordinator, 25, 365–371.
136. Phillips, D. A. (1992). Child care and parental well-being: Bringing quality of care into the picture. In A. Booth (Ed.), Child care in the 1990s: Trends and consequences (pp. 172–180). Hillsdale, NJ: Erlbaum.
137. Phillips, D. A., & Adams, G. (2001). Child care and our youngest children. The Future of Children, 11, 35–52.
138. Phillips, D. A., Howes, C., & Whitebook, M. (1992). Child care as an adult work environment. Journal of Social Issues, 47, 49–70.
139. Phillips, D. A., Lande, J., & Goldberg, M. (1990). The state of child care regulations: A comparative analysis. Early Child-hood Research Quarterly, 5, 151–179.
140. Piaget, J. (1952). The origins of intelligence in children. New York: International Universities Press.
141. Pianta, P. C., & Nimetz, S. L. (1991). Relationship between children and teachers: Associations with classroom and home behavior. Journal of Applied Developmental Psychology, 12, 379–393.
142. Pleck, E. H., & Pleck, J. H. (1997). Fatherhood ideals in the United Sates: Historical dimensions. In M. E. Lamb (Ed.). The role of the father in child development (3rd ed., pp. 33–48). New York: Wiley.
143. Power, T. G. (1981). Sex typing in infancy: The role of the father. Infant Mental Health Journal, 2, 226–240.
144. Power,T. G., McGrath, M. P., Hughes, S. O., & Manire, S. H. (1994). Compliance and self-assertion: Young children’s responses to mothers versus fathers. Developmental Psychology, 30, 980–989.
145. Ragozin, A. Z. (1980). Attachment behavior of day-care children: Naturalistic and laboratory observations. Child Development, 511, 409–415.
146. Raikes, H. (1993). Relationship duration in infant care: Time with a high ability teacher and infant-teacher attachment. Early Childhood Research Quarterly, 8, 309–325.
147. Raikes, H., Kisker, E., Paulsell, D., & Love, J. (2000). Early Head StartNationalResearchandEvaluationproject:Meetingthechild care needs of families. National Head Start Bulletin, 69, 7–10.
148. Raikes, H. H., & Love, J. M. (2002). History and purpose of Early Head Start. Infant Mental Health Journal, 23, 1–13.
149. Ramey, C. T. (2000). Michigan Family Impact Seminars: Child care and education. Michigan State University, East Lansing: Institute for Children, Youth and Families.
150. Ramey, C. T., & Campbell, F. A. (1984). Preventive education for high-risk children: Cognitive consequences of the Carolina Abecedarian Project. American Journal of Mental Deficiency, 88, 515–523.
151. Ramey, C. T., & Campbell, F. A. (1987). The Carolina Abecedarian Project: An educational experiment concerning human malleability. In S. S. Gallagher & C. T. Ramey (Eds.), The malleability of children (pp. 127–139): Baltimore, Brooks.
152. Ramey, C. T., & Ramey, S. L. (1998). Early intervention and early experience. American Psychologist, 58, 109–120.
153. Rohner, R. P. (1998, October). Father love and child development: History and current evidence. Current Directions in Psychological Science, 157–161.
154. Robinson, H. B., & Robinson, N. M. (1971). Longitudinal development of very young children in a comprehensive day care program:Thefirsttwoyears.ChildDevelopment,42,1673–1684.
155. Robinson, J. L., & Fitzgerald, H. E. (2002). Early Head Start: Contemporary perspectives and promise. Infant Mental Health Journal, 23, 250–257.
156. Rubenstein, J. L., & Howes, C. (1979). Caregiving and infant behavior in day care an in homes. Developmental Psychology, 15, 1–24.
157. Rubenstein, J. L., Pedersen, F. A., & Yarrow, L. J. (1977). What happens when mothers are away: A comparison of mothers and substitute caregivers. Developmental Psychology, 13, 529–530.
158. Rutter, M. (1976). Parent-child separation: Psychological effects on the children. In A. M. Clarke & A. D. B. Clarke (Eds.), Early experience: Myth and evidence (pp. 153–186). New York: Free Press.
159. Rutter, M. (1979). Maternal deprivation, 1972–1978: New findings, new concepts, new approaches. Child Development, 50, 283–305.
160. Rutter, M. (1995). Maternal deprivation. In M. H. Bornstein (Ed.), Handbook of parenting: Vol. 4. Applied and practical parenting (pp. 3–31). Mahwah, NJ: Erlbaum.
161. Salmi, M. (1994). The parental leave and day care systems in Finland. Helsinki, Finland: National Research and Development Centre for Welfare and Health.
162. Sameroff, A. J. (2000). Ecological perspectives on developmental risk. In J. D. Osofsky & H. E. Fitzgerald (Eds.), WAIMH handbook of infant mental health: Vol. 4. Infant mental health in groups at high risk (pp. 1–34). New York: Wiley.
163. Scarr, S., & Eisenberg, M. (1993). Child care research: Issues, perspectives, and results. Annual Review of Psychology, 44, 613–644.
164. Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory of genotype-environment effects. Child Development, 54, 424–435.
165. Shears, J., Robinson, J. L., & Emde, R. N. (2002). Fathering relationships and their associations with juvenile delinquency. Infant Mental Health Journal, 23, 79–87.
166. Singer, J. D., Fuller, B., Keiley, M. K., & Wolf, A. (1998). Early child-care selection: variation by geographic location, maternal characteristics, and family structure. Developmental Psychology, 34, 1129–1144.
167. Skeels, H. M. (1936). Mental development of children in foster homes. Journal of Genetic Psychology, 49, 91–106.
168. Smith, K. (2000). Who’s minding the kids? Child care arrangements: Fall 1995 (Current Population Reports, P 70-70). Washington, DC: U.S. Census Bureau.
169. Spitz, R. (1965). The first year of life. New York: International Universities Press.
170. Spitz, R. & Wolf, K. M. (1946). Anaclitic depression: An inquiry into the genesis of psychiatric conditions in early childhood: Vol. 2. The psychoanalytic study of the child (pp. 313–342). New York: International Universities Press.
171. Strelau, J. (1998). Temperament: A psychological perspective. New York: Plenum Press.
172. Summers, J. A., Raikes, H., Butler, J., Spicer, P., Pan, B., Shaw, S., Langager, M., McAllister, C., & Johnson, M. K. (1999). Lowincome fathers’ and mother’s perceptions of the father role: A qualitative study in four Early Head Start communities. Infant Mental Health Journal, 20, 291–304.
173. Suwalsky, J. T. D., Klein, R. P., Zaslow, M. J., Rabinovich, B. A., & Gist, N. F. (1987). Dimensions of naturally occurring motherinfant separations during the first year of life. Infant Mental Health Journal, 8, 3–18.
174. Tableman, B. (2001). Effective home visiting for very young children (Best Practice Brief No. 17). East Lansing: Michigan State University, University Outreach Partnerships.
175. Thelen, E., & Smith, L. B. (1994). Adynamic systems approach to the developmentofcognitionandaction.Cambridge,MA:MITPress.
176. Thomas, A., & Chess, S. (1977). Temperament and development. New York: Brunner/Mazel.
177. Thomas, A., & Chess, S. (1986). The New York Longitudinal Study: from infancy to early adult life. In R. Plomin & J. Dunn (Eds.), The study of temperament: Changes, continuities, and challenges (pp. 39–52). Hillsdale, NJ: Erlbaum.
178. Thomas, A., Chess, S., & Birch, H. G. (1968). Temperament and behavior disorders in children. New York: New York University Press.
179. Tronick, E. Z, Winn, S., & Morelli, G. A. (1985). Multiple caretaking in the context of human evolution: Why don’t the Efe know the Western prescription for child care. In M. Reite & T. Field (Eds.), Psychobiology of attachment (pp. 293–322). New York: Academic Press.
180. van Ijzendoorn, M. H., & De Wolff M. (1997). In search of the absent father: Meta-analysis of infant-father attachment. A rejoinder to our discussants. Child Development, 68, 604–609.
181. Wagner, M. G., & Wagner, M. M. (1971). Day care programs in Denmark and Czechoslovakia. In E. H. Grotberg (Ed.), Day care: Resources for decisions (pp. 28–37). Washington DC: United States Public Health Service, Office of Economic Opportunity.
182. Werner, E. E. (1988). A cross-cultural perspective on infancy. Journal of Cross-Cultural Psychology, 19, 96–113.
183. Wilson, D. B., & Chipaungu, S. S. (Eds.). (1996) Kinship care. Child Welfare, LSSV, 387–662.
184. Windle, M. (1991). The difficult temperament in adolescence: Associations with substance use, family support, and problem behaviors. Journal of Clinical Psychology, 47, 310–315.
185. Wong, M. M., Zucker, R. A., Puttler, L. I., & Fitzgerald, H. E. (1999). Heterogeneity of risk aggregation for alcohol problems between early and middle childhood. Development and Psychopathology, 11, 727–744.
186. Young, K. R., Marsland, K. W., & Zigler, E. (1997). The regulatory status of center-based infant and toddler child care. American Journal of Orthopsychiatry, 67, 535–544.
187. Yoshikawa, H. (1994). Prevention as cumulative protection: Effects of early family support and education on chronic delinquency and its risks. Psychological Bulletin, 115, 28–54.
188. Yoshikawa, H. (1995). Long term effects of early childhood programs on social outcomes and delinquency. The Future of Children, 5, 51–75.
189. Yogman, M. W. (1982). Development of the father-infant relationship. In H. E. Fitzgerald, B. M. Lester, & M. W. Yogman (Eds.), Theory and research in behavioral pediatrics (pp. 221–279). New York: Garland Press.
190. Zucker, R. A., Fitzgerald, H. E., Refior, S. K., Puttler, L. I., Pallas, D. M., & Ellis, D. A. (2000). The clinical and social ecology of children of alcoholics: Description of a study and implications for a differentiated social policy. In H. E. Fitzgerald, B. M. Lester, & B. S. Zuckerman (Eds.), Children of addiction: Research, health and policy issues (pp. 109–141). New York: Garland Press.
191. Zukow-Goldring, P. (1995). Sibling caregiving. In. M. H. Bornstein (Ed.), Handbook of parenting: Vol. 3. Status and social conditions of parenting (pp. 177–208). Mahwah, NJ: Erlbaum.