|
Article Excerpt
Sensitive and responsive caregiving is associated with better cognitive and language outcomes. Using the longitudinal data set from the National Institute for Child Health and Human Development (NICHD) Study of Early Child Care and Youth Development, this study asks how changes in the sensitivity of both mothers and caregivers from 6 months to 6 years relates to language and academic outcomes at the start of formal schooling. Three questions are posed: (1) How variable is the quality of caregiving that children experience from mothers and child care providers during early childhood? (2) Do children benefit from both sensitive parents and sensitive caregivers? (3) Are changes in sensitivity over time related to cognitive and language outcomes at the end of preschool and the beginning of formal education? Person-centered and variable-centered analyses revealed that children experience changing patterns of sensitivity across time, that children benefit from sensitive interactions with all adults, and that changes in the sensitivity children experience across time are associated with both language and cognitive outcomes.
**********
Stimulating and responsive caregiving environments promote social and cognitive development (Shonkoff & Philips, 2000). A substantial body of research attests to this claim (see Bornstein & Tamis-LeMonda, 1989, and Tamis-LeMonda & Bornstein, 2002, for reviews). Adults who take turns in interactions with young children, share periods of joint focus, and express positive affect provide infants and toddlers with a secure base for exploring their world and with the scaffolding needed to facilitate language and cognitive growth (Bradley et al., 1989; Bronfenbrenner & Morris, 1998; Clarke-Stewart, 1973; Howes, 2000; Katz, 2000; Tomasello & Farrar, 1986). For example, evidence strongly suggests that a greater amount and a larger diversity of verbal stimulation foster earlier and richer language outcomes in terms of both vocabulary and grammar (Beebe, Jaffe, & Lachman, 1992; Hart & Risley, 1995, 1999; Huttenlocher, Haight, Bryk, Seltzer, & Lyons, 1991; Snow, 1986; Tamis-LeMonda, Bornstein, & Baumwell, 2001; see Hoff, in progress, for a review).
There is little debate that sensitive and stimulating caregiving plays an important role in promoting cognitive and language development. Most research, however, treats stimulation and responsiveness as a static variable and asks how characteristics of the mother-child or caregiver-child interaction at one point in time or averaged over time affects either concurrent or later child behavior. Little is known about how individual variation in parental or caregiver sensitivity over time affects child outcomes or even whether parental and caregiver sensitivity is constant over time (but see Bornstein & Tamis-LeMonda, 1989; Landry, Smith, Swank, Assel, & Vellet, 2001; Landry, Smith, Swank, & Miller-Loncar, 2000; Tamis-LeMonda & Bornstein, 2002). This study uses the longitudinal data set from the NICHD Study of Early Child Care and Youth Development to examine this issue. It investigates (1) the degree to which children from 6 to 54 months receive consistent or inconsistent patterns of stimulation and responsiveness from parents and changing child care providers, and (2) the extent to which patterns of change in the responsiveness children receive over time predicts their development at the start of formal schooling. Two statistical procedures not widely used in the cognitive literature are used to address these questions.
Sensitive and stimulating parenting. Stimulating and responsive parenting in early childhood is one of the strongest predictors of children's later language, cognitive, and social skills (Bronfenbrenner & Morris, 1998; Sameroff & Seifer, 1983). Distal measures show that homes rich in age-appropriate toys, reading materials, and conversation provide opportunities for learning that consistently predict higher reading and math scores during elementary and middle school (Bradley & Corwyn, 2002; Bradley et al., 1989; Bradley, Corwyn, Burchinal, Pipes McAdoo, & Garcia Coll, 2001; Senechal & LeFevre, 2002; Whitehurst, Arnold, Epstein, & Angell, 1994; among many others), reading competency in second grade (Scarborough, Dobrich, & Hager, 1991), and even 11th-grade reading comprehension (Cunningham & Stanovich, 1997; Cunningham, Stanovich, & West, 1994). Children's language skills are even more strongly related to proximal measures of quality in parent-child interaction, such as sensitivity cooperation, acceptance, and responsiveness (Landry et al., 2001; Tamis-LeMonda & Bornstein, 2002; Wakschlag & Hans, 1999). Parental warmth demonstrated as open displays of affection, physical or verbal reinforcement, and sensitivity to children's requests and feelings are also significantly associated with academic achievement and cognitive growth (Bornstein & Tamis-LeMonda, 1989; Burchinal, Campbell, Bryant, Wasik, & Ramey, 1997; Howes, Phillips, & Whitebook, 1992; Landry et al., 2001; Morrison & Cooney, 2002). The affective quality of mother-child interactions in early childhood is further related to early cognitive competencies, such as mental ability scores at age 4, school readiness skills at age 5 and 6, IQ scores at age 6, and vocabulary and mathematics performance at age 12 (Estrada, Arsenio, Hess, & Holloway, 1987). Responsiveness of parents in terms of diversity of language also relates to later proficiency (Weizman & Snow, 2001).
Sensitivity and stimulation in child care. Although the role of sensitive input has been more extensively explored in the parenting literature, responsive and stimulating behavior by caregivers also relates independently to child outcomes. At the distal level, all large, multi-site, observational studies of children's cognitive and language development (Howes et al., 1992; Love et al., 2003; NICHD Early Child Care Research Network [ECCRN], 2000, 2002; Peisner-Feinberg & Burchinal, 1997; Zill, 1999) and most smaller studies (Burchinal, Roberts, Nabors, & Bryant, 1996; Burchinal et al., 2000; Dunn, 1993; Kontos, 1991; McCartney, 1984; Schliecker, White, & Jacobs, 1991) find a direct relationship between the amount of sensitivity in the environment and cognitive and language outcomes. This link between child care quality and child outcomes has been observed in child care homes and relative care as well as in center care (Clarke-Stewart, Vandell, Burchinal, O'Brien, & McCartney, 2002; Kontos, Howes, Shinn, & Galinsky, 1997; NICHD ECCRN, 2000, 2002).
In language development, several studies looked more proximally at the caregiver's interaction with the target child. Children whose caregivers were more responsive and stimulating showed higher language scores and larger gains over time (Burchinal et al., 2000). Studies also find direct relationships among responsiveness, stimulation, and language outcomes (McCartney, 1984; NICHD ECCRN, 2000).
The research relating parental and child care interaction with child outcomes has provided powerful information on environmental predictors of cognitive and language success. Yet, research exploring this question generally relies upon group data in which the predictors of stimulation and responsiveness are measured at only one point in time (Hart & Risley, 1995; Hoff & Naigles, 2002; Huttenlocher et al., 1991 ; Newport, Gleitman, & Gleitman, 1977), are averaged over time (see NICHD ECCRN, 2000, and Tamis-LeMonda & Bornstein 2002, for a review of their work), or are treated as a time-varying covariate (Burchinal et al., 1997; NICHD ECCRN, 2004). Although these studies are informative, they generally mask individual and group variation that might accrue across time. These investigations do not allow us to ask whether children are receiving constant amounts of sensitivity over time from the parents and caregivers or whether changes in stimulation and sensitivity that they receive over time have any notable effects on child outcomes. That is, they do not treat the sensitivity and stimulation that children receive as a dynamic variable.
Limited evidence suggests that changes in parental sensitivity over time relate to changes in child outcomes (see Bornstein & Tamis-LeMonda, 1989; Landry et al., 2000, 2001; NICHD ECCRN & Duncan, 2003; Tamis-LeMonda & Bornstein, 2002). Landry et al. (2001), for example, found that children who had highly sensitive parents in the first 3 years of life followed by lower sensitivity did not perform as well as children who had consistently high sensitivity across early childhood. Bornstein and Tamis-LeMonda (1989) also looked at patterns of sensitivity and stimulation over time. A longitudinal study of 40 mother-child dyads when the children were between 9 months and 19 months of age revealed that prompt, contingent, and appropriate responses to children's behaviors had dramatic effects on children's later language and cognitive growth. In both the Landry et al. (2001) and the Bornstein and Tamis-LeMonda (1989) studies, more sensitive parents were more stimulating. Operational definitions of sensitivity often embed qualities of stimulation.
The current study expands upon these longitudinal investigations in three ways. First, studies of patterns of change in responsiveness over time and its relationship to cognitive and language outcomes focus almost exclusively on sensitivity/stimulation as it is experienced in the parent-child interaction. With a large proportion of children in alternative care, it is important to ask how sensitivity that emerges in both parent and caregiver settings relates (independently and jointly) to child outcomes. Second, available longitudinal studies look at development only until 4 years of age. This study pushes the developmental envelope by asking how patterns of change in sensitivity and responsiveness from mothers and caregivers across time affects cognitive and language outcomes at the end of preschool and after the transition to school in first grade. Finally, this study introduces two relatively new analytic techniques to address questions about relations between sensitivity and responsiveness and child outcome. The variable-centered approach estimates changes in sensitivity and responsiveness in terms of individual growth curves and uses individual differences in growth curve parameters as predictors of child outcomes. In contrast, the person-centered approach identifies types of patterns in the change in sensitivity/stimulation over time, using types as predictors that might differentially relate to outcomes.
Methods
Participants
The NICHD Study of Early Child Care and Youth Development offers a unique opportunity to address these issues with a large geographically, ethnically, and economically diverse population. Children born in hospitals at 10 geographic sites in the United States were followed from birth to first grade. Families were recruited shortly after the child's birth in 1991. For a full description of the sample recruitment and distribution see NICHD ECCRN (2004). A total of 1,097 of the 1,364 original participants continued in the study through 54 months and form the sample for the current study. Mothers of the children in this sample had an average of 14.2 years of education (SD = 2.51); 18% were single; and average family income was about four times the poverty threshold (M = 3.7, SD = 2.74). About three-fourths of the children were European American, non-Hispanic. Importantly, although the analysis sample is not nationally normed, it reflects a diverse range of family backgrounds present in the United States.
Procedure
Infants and their mothers were seen in the lab or in their homes repeatedly from the time the child was 1 month of age until the child was in first grade. Mothers also responded to questions in a telephone interview that was given every 3 months up until age 36 months and approximately every 4 months thereafter. Data for this study were collected using multiple methods: standardized observations of the child or the mother and child, telephone interviews for the mother, and lab-based standardized tests for child outcome measures. Below we outline the measures that were used to provide demographic data, the predictor variable of sensitivity from both mothers and caregivers, and child outcome variables in language, attention, and academic achievement at 54 months of age and in first grade. These outcome measures were collected at the end of preschool (54 months) and in the spring of their first grade.
Measures
PREDICTOR VARIABLES
Demographic measures. During home interviews at 1 month, mothers reported their education (in years) and the study children's sex and ethnicity (non-Hispanic African American, non-Hispanic European American, Hispanic, or other). The presence of a husband or partner in the...
|
