2007 Society for Research in Child Development Biennial Meeting Abstracts

“K-5 Reading Achievement Trajectories and Children's Home Environment: Head Start Graduates in the ECLS-K” Jerry West, Louisa Banks Tarullo, and Nikki L. Aikens. Successive cohorts of a longitudinal descriptive study of Head Start children, the Head Start Family and Child Experiences Survey (FACES), show that most Head Start children have below-average literacy skills at entry and exit from the program (ACF, 2003). However, the highest score quartile of Head Start entrants scored at or above national averages in vocabulary and early writing, while the mean standard scores for the lowest quartile of Head Start entrants were at least two standard deviations below national averages. By the spring of kindergarten, the average child was approaching national norms in vocabulary and early writing, but children who started Head Start with scores in the highest quartile tended to show the least gains. These data lead to two questions: 1) What are the learning trajectories of these different groups of children from kindergarten through grade 5? and 2) What factors in children's home environment and later school experiences may influence variations in their learning trajectories through the early grades?

Data from the Early Childhood Longitudinal Study-Kindergarten Cohort (ECLS-K) are used to help answer these questions. The ECLS-K included a sample of 2,200 children for whom Head Start attendance was verified. The ECLS-K assessed children's reading knowledge and skills using a two-stage adaptive test. A parent interview at each round included a battery of items on literacy-related activities (e.g., reading to or with the child, taking the child to a library) and resources (e.g., number of books in the home, access to a public library and having a library card).

The sample of Head Start-verified children was divided into three groups, characterized by low, middle, and high scores on the ECLS-K reading achievement test at kindergarten entry. The top third entered kindergarten with reading scores several points above the national average, and maintained this trajectory through spring of third grade. Children in the high group were more likely to be white and more advantaged (e.g., higher household income and more home literacy resources) than children in the other two groups.

Preliminary analyses of the ECLS-K kindergarten through grade 3 using Hierarchical Linear Modeling techniques suggest that Head Start children in the high-achievement group not only had higher reading scores in the fall of kindergarten, but also gained reading skills at faster rates through third grade in comparison to the low- and middle-level groups. Differences in gains between low- and high-achievement groups were greater for children in households below poverty than for other children, and differences in the rates of growth between the three achievement groups were smaller for minority children than for white children. Evidence suggests that home literacy resources may moderate the relationship between family poverty and children's reading achievement gains. This paper will examine in greater depth the relationships between literacy-activities and resources and children's K-5 reading achievement trajectories for the three initial achievement groups.

“The Fathering Effect: Minority Fathers' Contributions to Their Young Children's Development” Jacqueline D. Shannon, Natasha Cabrera, Jerry West, and Karen McFadden. The growing numbers of nonresident fathers and social fathers, along with increased maternal labor force participation have altered fathers' roles in families. Although recent research has focused on the roles of fathers, few have included minority fathers with young children. The majority of research on fathers has focused on two-parent, predominately white families, and has often used a present/absent definition of father involvement. Consequently, we know relatively little about how minority fathers are involved with their young children, and about the antecedents and consequences of father involvement within a racially/ethnically diverse group of families. This oversight is noticeable when we consider the increased diversity of the U.S. population. More than a quarter of the U.S. population is now composed of racial/ethnic minorities, including African Americans, Latinos, and Asian Americans.

To address this gap, we use the Early Childhood Longitudinal Study—Birth Cohort data, the first nationally representative longitudinal study of young children that includes a large number of minority children, to answer the following research questions: 1) How often and in what ways are minority fathers involved (accessible, responsible, engaged) with their children at 9 and 24 months? 2) How do fathers' human and financial capital (education, income, employment) and children's gender relate to father involvement? 3) What are the independent effects of these characteristics and involvement on children's development at both time points? 4) What are the moderating effects of fathers' race/ethnicity on the relationship between fathers' involvement and children's outcomes?

The sample in our study is comprised of 4,287 biological mothers and 2,992 biological fathers from three primary racial/ethnic backgrounds (African American, Latino, and Asian American). Preliminary findings indicate that 84 percent of all 9-month-old infants born in 2001 have access to their biological fathers, with 95 percent being Asian American, 89 percent being Latino, and 72 percent being African American children seeing their fathers daily. These accessible fathers take responsibility for their children, although less than a third signed a legal or informal paternity agreement. They also report that they regularly engage in a range of activities with their children; however, they engage in more caregiving than physical play or literacy activities. Although African American fathers are less accessible, they report being engaged in more caregiving than other minority fathers. As a group, fathers engaged in more activities when mothers were employed. They engaged in more literacy and physical contact with sons and more caregiving with daughters. Also, maternal employment and fathers' literacy predicted infants' 9-month cognitive scores. Further analyses will examine these pathways separately for each minority group and longitudinally. Findings will inform educators, researchers, and policymakers about the ways in which minority fathers are involved in the lives of their young children and how their involvement contributes to their children's development across multiple domains.

"Impacts of Early Head Start at the End of the Program (Age 3) and Two Years Later When Children Were in Prekindergarten" John M. Love, Cheri Vogel, Helen H. Raikes, Rachel Chazan-Cohen, Ellen E. Kisker, Jill Constantine, Welmoet vanKammen, and Jerry West. For 10 years, the Early Head Start Research and Evaluation Project has studied a sample of 3,001 children and their low-income families who were randomly assigned to receive EHS services or to a control group in 17 communities around the U.S. After leaving EHS, many children participated in Head Start, prekindergarten programs, and other formal programs between ages 3 and 5. Many control group children also entered formal programs after age 3, but EHS increased children's enrollment rates in Head Start and other formal programs relative to the control group. This poster summarizes the age-3 impacts, impacts of EHS at the prekindergarten period, and plans for the fifth-grade followup. It also sets the stage for other posters in the series.

A number of the many modest impacts on children and parenting seen at age 3 (reported previously but these will be summarized) were still present when children were about age 5. Two years after the EHS program ended, EHS significantly reduced children's behavior problems (p<.05, effect size=.10) and enhanced positive social skills and approaches to learning (p<.01, ES=.12). There were no impacts on vocabulary (except for Spanish-speaking children) or achievement-related outcomes. There were no longer impacts on videotaped parent-child interaction outcomes. However, EHS continued to have significant impacts on parents' support for children's learning (e.g., daily reading [p<.05, ES=.09] and the home environment [p<.01, ES=.13]), and a new benefit emerged for parents—a reduction in their risk of depression (p<.05, ES=.10).

However, the patterns of impacts for subgroups based on family and program characteristics are somewhat different from those at age 3, particularly for high-risk families and children and families from home-based programs, as described in posters 4 and 5, or strikingly similar (poster 6). In addition, nonexperimental analyses of the contribution of children's formal program experiences after EHS demonstrated the combined effects of experiences before and after age 3.

Starting in winter 2007, the Fifth-Grade Follow-Up Study will provide new information about the children as many prepare to enter middle school. The follow-up data will be used to (1) estimate the effects of EHS on child and family service receipt and outcomes, (2) describe the elementary school experiences of children who received EHS in contrast to the control group, and (3) examine the relationships between children's elementary school experiences—alone and in combination with their preschool experiences—and child and family outcomes.

The follow-up design includes a home visit and a teacher questionnaire, and school records. The home visit includes (1) child assessments, (2) child interviews, (3) observations, and (4) maternal interviews. The child assessment consists of the Peabody Picture Vocabulary Test -III and the fifth-grade reading and math assessments from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-1999, all administered using a computer-assisted interview mode. Children's height and weight are measured also. Other data include mother-child interactions, information about children's school experiences, and maternal ratings of children's social skills and problem behaviors.

"Age 3 Mediators of Early Head Start Impacts and Prekindergarten Outcomes" Mary Maguire Klute, Rachel Chazan-Cohen, Helen H. Raikes, and Ellen E. Kisker. This poster explores the question: Do age 3 outcomes mediate the relation between EHS and prekindergarten outcomes? In sum, the answer is yes. EHS had impacts on children's functioning at age 3 (notably cognitive functioning as assessed by the Bayley MDI and children's engagement of their parent in play) that mediated cognitive and social-emotional outcomes at kindergarten entry.

We examined seven mediators covering a range of outcomes observed at age 3 as well as out-of-home experiences between age 3 and 5. Age 3 outcomes included: children's cognitive and language abilities at 36 months (Bayley MDI and PPVT); children's behavior problems at 36 months (CBCL-ASEBA); parent-child interactions (child's engagement of the mother during play); home environment (HOME). Children's out-of-home experiences between age 3 and 5 included whether they were enrolled in formal care at ages 3-4 and 4-5 and ever enrolled in Head Start (HS).

Age 5 outcomes included: social-emotional measures (parent-reported child behavior problems and examiner-rated Leiter Cognitive Social), and cognitive/language outcomes (Woodcock Johnson Letter-Word Identification [LWI] and Leiter Attention Sustained). Structural equation models using Mplus were used; we used missing data features of Mplus to use all available data. We examined the significance of indirect effects to determine whether there was significant evidence for mediation. The models found that EHS is significantly and positively related to age 3 HOME, Bayley, and child engagement in play as well as Head Start participation between age 3 and 5.

Results supported the following mediated pathways:

•  Child engagement of parent during play at age 3 mediated the relationship between EHS and behavior problems such that children assigned to EHS engaged their mothers more in play at age 3 and this, in turn, was related to fewer parent-reported behavior problems at kindergarten entry.

•  Both Bayley MDI and child engagement of parent during play mediated relations between EHS and Leiter Cognitive-Social observation ratings at kindergarten entry.

•  Bayley MDI at age 3 mediated the relationship between EHS and Woodcock Johnson LWI. EHS children had higher Bayley scores at age 3, and Bayley scores were in turn related to higher LWI scores at 5. There was also a trend for age 3 HOME scores to mediate LWI scores.

•  Bayley MDI scores and child engagement during play at age 3 mediated the relation between EHS and Leiter Sustained Attention at kindergarten entry.

In conclusion, EHS program involvement through children's cognitive abilities at age 3 was indirectly related to both positive social-emotional and cognitive outcomes at prekindergarten, specifically Leiter Cognitive Social, Woodcock-Johnson Letter-Word Identification, and Leiter Sustained Attention. Additionally, children's ability to engage their mothers in play at age 3 mediated the relationship between EHS and both social-emotional and cognitive outcomes at prekindergarten, specifically problem behaviors, Leiter Cognitive-Social, Letter-Word Identification, and Sustained Attention (trend).

"Influences of Early In-Home and Out-of-Home Environments on Child Functioning at Kindergarten Entry" Rachel Chazan-Cohen, Helen H. Raikes, Ellen E. Kisker, Catherine Ayoub, Lori A. Roggman, Barbara Alexander Pan, Allison Sidle Fuligni, and Jeanne Brooks-Gunn. The Early Head Start (EHS) Research and Evaluation Project assessed the contexts of early child development, including family and parent-child relationship factors as well as aspects of the early care and education settings children experience. This paper will examine the role of in-home experiences as well as out-of-home experiences over time on children's functioning in the spring prior to kindergarten entry. While the primary goal of the EHS Research and Evaluation Project was to assess the role of EHS and other services on child and family outcomes, we are able to ask additional questions, such as: How stable are environmental contexts for low-income children? How are these early contextual factors related to later child functioning?

In-home experiences include three aspects of the family context: support for learning and growth observed in the home environment (HOME scale), parent sensitivity in a videotaped free-play task (NICHD three-bag task), and aspects of maternal mental health (CES-D, Parenting Stress Inventory, and family conflict subscale of the Family Environment Scale). These factors were assessed when children were 14, 24, and 36 months of age, and again before children were to enter kindergarten. We examined the influence of 14-month home environment as well as change over time. Out-of-home experiences included EHS or control group 0-3 and use of formal care 3-5 (formal care could be center-based care, state prekindergarten, or Head Start). In this poster we present new findings from analyses of in-home and out-of-home experiences in the same regression models to learn about relative contributions of these two contextual influences on low-income children's development.

Previous analyses showed that early parenting and home environments predict child outcomes at kindergarten entry, and furthermore, that while family environments are somewhat stable over time, there is meaningful change over time. More optimal parenting and home environments at 14-months were associated with more optimal child outcomes, although change over time was also important—positive change was associated with more optimal outcomes, and negative change was associated with more negative outcomes. Children's vocabulary and behavior problems were associated with all aspects of parenting and the home-environment, but pre-academic skills were most strongly associated with the HOME scale. In terms of out-of-home experiences, 2 years after the end of the program, EHS continued to benefit children in social-emotional outcomes, while formal care 3-5 was associated with improved pre-academic skills and more aggressive behaviors.

Preliminary regression analyses looking at the combined effect of in-home and out-of-home environments show that both contribute in an additive way to child functioning at kindergarten entry. While the betas are larger for the home environment factors, and the home-environment is associated with all aspects of child outcomes, the out-of-home factors continue to exert influence on children above these powerful influences (see also, NICHD Early Child Care Research Network, 2005). Implications for practice and research will be explored.

"Outcomes for Children in Families at Highest Risk at Ages 3 and 5" Ellen E. Kisker, Helen H. Raikes, Rachel Chazan-Cohen, Judith Carta, Catherine Ayoub, and Jini Puma. Families defined at high risk in EHS (i.e., with four or five demographic risk factors in addition to living in poverty), was the only subgroup of families studied for whom there were no positive impacts of EHS when children were 3. By age 5, however, some important positive impacts of EHS had emerged. This poster explores what happened over this period and suggests implications for programs working with high-risk families. The five risk factors, measured when families entered the study included (1) mother was a teen parent, (2) did not complete high school, (3) was unmarried, (4) received AFDC/TANF, and (5) was not enrolled in school or employed. We created three risk levels for analysis: (1) low (0-2 risks), (2) moderate (3 risks), and (3) high (4-5 risks). About 25 percent of families were high risk. This poster examines reasons for lack of favorable impacts at age 3, reports impacts that emerged at age 5, examines children's experiences in formal care and education programs between ages 3-5, and reports the combined effects from EHS and Head Start in this group at age 5.

In investigating the lack of impacts for the high-risk group at age 3, we found that these EHS parents were difficult to engage and participated less consistently in services than was true for other parents. However, a small subgroup of highest risk children and families who did engage in services at an intensive level demonstrated favorable outcomes relative to other high-risk children and families, suggesting that lack of favorable outcomes may be largely due to low or erratic family involvement in program services. At age 5, several positive impacts emerged: EHS children in the high-risk group had significantly better approaches to learning as reported by their parents, although there was a negative impact on letter-word identification. Parents were observed to provide more support and stimulation than was true for control group parents. EHS children and parents were significantly less likely to be living in environments where they witnessed violence and to be living with someone with an alcohol or drug problem than control group children. Parents were less likely to report having been physically, emotionally, or sexually abused in the past year.

Unlike other subgroups, EHS did not increase attendance in formal care and education programs between ages 3 and 5 for children in high-risk families and highest-risk children were less likely than other groups to be in formal care (39 percent vs. 43 percent for medium, and 47 percent for low-risk groups). However, attending Head Start, but not other formal programs, was associated with favorable age 5 outcomes in children's cognitive development, parental home environments, reading to children, and in reduced parental depression. The study concludes that EHS programs may be changing conditions for families when children are 3 that translate into impacts by age 5, and that comprehensive services for children throughout the period from birth to age 5 appear to be important for children in families at highest risk.

"The Impact of Home-Based Programs on Children and Families in Early Head Start" Anne Martin, Rachel Chazan-Cohen, Jeanne Brooks-Gunn, Helen H. Raikes, John M. Love, Ellen E. Kisker, Cheri Vogel, and Brenda Jones-Harden. EHS programs were classified according to their service delivery model: home-based (n = 7), center-based (n = 4), or mixed (which provided home-based services to some families and center-based services to others, or provided both types of services to all families; n = 6). Here we focus on impacts on children and families in home-based programs.

Consistent with the programs' theories of change, home-based programs tend to have impacts on parent outcomes. At 36 months, when services ended, home-based programs had beneficial impacts on the home environment (HOME scores) and on parent-child interaction. EHS children were more able to engage their parents in play and parents were more supportive during video-taped semi-structured play sessions. Parents reported less parenting stress than control group parents. These parents were more likely to be in school than their control group counterparts. Furthermore, when home-based programs fully implemented the Head Start Program Performance Standards (n = 4), they had a broader pattern of significant impacts, including positive impacts for children's cognitive and language development. Parents in these sites spanked less, reported less parenting stress than the control parents, and were more likely to have attended high school. Parents in less fully implemented home-based programs (n = 3) had positive impacts on social-emotional development (engagement and attention) and parent education.

At age 5, when children were preparing to enter kindergarten, positive impacts persisted for children and families who had been in home-based programs. Compared to the control group, children who had been in EHS showed reduced hyperactivity, behavior problems, and withdrawn behavior, as well as more positive social skills and approaches to learning and higher Woodcock-Johnson Applied Problems scores. Their parents were more likely to read daily, and they provided more optimal home environments (i.e., scored higher on the HOME total, learning environment, and warmth scales). Their households also had higher monthly incomes ($2,408 vs. $2,106), a finding that suggests a possible economic benefit of EHS. EHS participants also showed greater use of formal care arrangements (46 percent vs. 39 percent) and Head Start participation (57 percent vs. 51 percent) between the ages of 3 and 5.

Further analyses of the home-based group will examine whether treatment impacts on children and parents at 36 months, as well as formal child care between ages 3 and 5, mediates the treatment impacts found at age 5.

"Impacts of Early Head Start on African American Children at Ages 3 and 5" Helen H. Raikes, John M. Love, Rachel Chazan-Cohen, Cheri Vogel, and Ellen E. Kisker. Substantial program impacts of Early Head Start were found at age 3 for children in African American families, including favorable impacts on children's sustained attention with objects (effect size [ES] = .48) and their engagement of parents during play (ES = .48), and reductions in child negativity toward parent (ES = . 37) and parent-reported aggressive behavior problems (ES = .35). Positive impacts were found for children's receptive vocabulary (ES = .23) as well as for a number of parenting outcomes, including supportiveness during play (ES = .47), home environment (HOME total, ES = .27); reduced parenting stress (ES = .20) and parent-child dysfunctional interaction (ES = .23); and adherence to a regular bedtime (ES = .23). As a result of EHS participation, African American parents were more likely ever to have been employed since enrollment (ES = .28) and spent more hours per week in education or training (ES = .28); they received less in welfare cash assistance (ES = .20).

At age 5, the African American subgroup continued to show substantial impacts from involvement in the Early Head Start program. The treatment group had significantly higher vocabulary scores (ES = .20); book knowledge and comprehension (ES = .21); attention (ES = .24); positive approaches to learning (ES = .26); and fewer parent-reported aggressive behavior problems (ES = .20). Parents displayed greater cognitive stimulation during videotaped play (ES = .21) and less negative regard (ES = .43). The parents owned more children's books (ES = .24). These parents were less likely to have lived with someone who had a drug or alcohol problem (ES = .30), to have been abused in the past year (ES = .23), or to be depressed (ES = .24) than parents in the control group.

African American children were highly likely to access formal care and education during the post-EHS period when they were 3 to 5 years of age; 93 percent of African American children were in formal care and education (more than any other subgroup) at some time between 3 and 5, with 53 percent attending some formal programs in both the 3-4 and 4-5 age periods; 64 percent never attended Head Start. Children who attended formal care and education ages 3-4 and 4-5 had significantly higher Woodcock Johnson Letter-Word Identification scores than African American children who did not attend formal care; a similar pattern was seen if they attended Head Start particularly and African American children who attended Head Start experienced an additional positive effect on HOME scores but no other effects were directly associated with formal care and education or Head Start. Sustained effects of Early Head Start among African American children are attributed both to Early Head Start before age 3 (as seen in the experimental analyses) and to high levels of involvement in formal care and Head Start that supported their early gains after age 3.

"Tangled Adversity and Early Academic Achievement: Understanding Effects from School-Level Poverty and Minority Segregation" Nikki L. Aikens and Kirsten Kainz. The social composition of schools first became a prominent issue in public education with the landmark Brown v. Board of Education decision of 1954. Since that time, research, led largely by the 1966 Coleman Report, has highlighted the relationship between the race/ethnicity and socioeconomic status of a child's peers to their own achievement. However, very little of this work has focused on very young children. This work also has evidenced mixed findings on whether school composition impacts ethnically diverse children similarly, and virtually none of it has examined the effects on Latino children. Using data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K), the present research highlights the effects of school racial and economic composition on children's early achievement and growth in reading and  math and how such effects differ by ethnicity.

Results indicate that poor and minority children are more likely to attend racially segregated and high-poverty schools in the kindergarten year than their peers. In addition, across children, attending a high-poverty school or school with high concentrations of minority students is associated with lower achievement scores in the fall of kindergarten, as well as constrained reading and math performance  through fifth grade. Comparative models across White, African American, and Latino children suggest that the effects of school composition are more strongly associated with White and Latino children's initial kindergarten reading and math achievement, while school composition constrains the rates of achievement growth for African American children.

Back to Top