Head Start and Residentially Mobile Young Children's Development

• In Event: 1-122 - Educational and Family Contexts of Residentially Mobile Children: Potential Pathways for Improving Child Outcomes

Integrative Statement

Almost one-quarter of young children moved in 2015 (US Census, 2016). Although typical, moving—especially multiple times—may undermine low-income children’s development. The early childhood years are an important period for addressing social inequalities in health and well-being related to poverty and residential mobility (Shonkoff et al., 2012). No existing policies promote highly mobile children’s development. Given their documented benefits for disadvantaged children and wide public support (e.g., Burchinal, Magnuson, Powell, & Hong, 2015), high quality public preschool programs, such as Head Start, may be an avenue for intervening to support low-income highly mobile children. The current paper explores whether Head Start participation moderates how mobility is associated with low-income children’s development, and potential family pathways that explain those associations.
This paper uses data from the Head Start Impact Study 2002-2006 (HSIS), a representative study of ~4,700 children of 3- and 4-year olds, randomly assigned to one of 320 Head Start centers or a comparison group. Children were followed annually from the beginning of the Head Start year through the end of kindergarten using a variety of assessments, interviews, and informants. This paper focuses on children identified as White, African American, or Hispanic (N = 3,725) and their mothers, with 2,433 assigned to Head Start and 1,292 to a comparison group (did not receive Head Start but could access other preschool options). Most children were low-income (<2 income-to-needs ratio), and 21% were White, 30% African American, and 27% Hispanic. Approximately 11% moving at least three times (Table 1). Mobility in the year prior to Head Start was categorized into low (1 move) and high (2 or more moves), with no moves as omitted referent. Indicators of overall child development at the end of kindergarten were captured through multi-component measures of socioemotional functioning and cognitive development.
Multilevel fixed effects models utilized regression-based techniques and instrumental variable analyses to determine how mobility, Head Start participation, and their interactions predicted children’s development. All models accounted for a robust set of individual, maternal, family, and center characteristics. Table 2 presents weighted descriptive statistics on children’s outcomes by Head Start treatment and mobility status. With a few exceptions, highly mobile children have worse outcomes than stable children across treatment conditions. Preliminary results suggest that Head Start moderates the negative association between high mobility and early achievement as assessed by WJ-III applied problems and letter-word scores (β = 40.43, and β = 69.33, respectively, p < .1), but not socioemotional functioning.
To help address unmeasured confounds tied to residential mobility that may undermine causal inferences drawn by HSIS, additional analyses will use propensity score matching methods. The final paper will explore associations with physical health outcomes, as well as the extent to which parenting behaviors explain potential associations. Results will help illuminate whether public preschool programs such as Head Start can support low-income, highly mobile children’s development, and explore possible pathways for how this support is achieved.

Authors

• Robert Julius Anastasio, Eliot-Pearson Department of Child Study and Human Development, Tufts University
  Presenting Author

• Tama Leventhal, Eliot-Pearson Department of Child Study and Human Development, Tufts University
  Non-Presenting Author

• Sara Anderson, West Virginia University
  Non-Presenting Author