A causally informed approach to examining the effects of dimension, dosage, and timing of SES adversity on parent-child interactions across the first years of life

• In Event: Poster Session 14
  In Poster Session: PS 14 Topic - Parenting & Parent-Child Relationships

Abstract

The dynamics of parent-child engagement represent the foundation for social life and are a critical context for early learning. Considerable theoretical and empirical work suggests that high-quality parent-child dynamics can be undermined by the litany of stressors experienced by families experiencing economic adversity (e.g., Bradley & Corwyn, 2002; Evans, 2004; Magnuson & Duncan, 2002). However, most existing literature has been limited by the use of broad distal proxies and cumulative summaries of socioeconomic resources as well as methods that preclude understanding of both the role of developmental timing and causal structure. To fill this gap, the current study employed an under-utilized causally informed technique to examine the relative influence of dimension, dosage, and timing of exposure to socio-economic adversity on parent-child interactions.

Data come from a prospective, population-based longitudinal sample of children (N = 1,292;48% female; 45% African American) living in rural low-income communities (Family Life Project; Vernon-Feagans et al., 2013). Repeated measures of sociodemographics and indices of maternal sensitivity and child positive affect (coded via a free play task) were collected at home visits when the children were 6, 15, 24, 35, and 58 months of age. Latent factor scores representing dimensions of socioeconomic adversity (material deprivation, sociocognitive resources, and psychosocial threat via parental conflict; DeJoseph et al, in press) were established for each time point.

Marginal structural models (MSMs; Robins et al., 2000) leverage inverse propensity of treatment weighting as a means of balancing potential confounds including time-invariant and time-varying control covariates and past adversity exposures. We established balancing weights for each time point of adversity exposure after identifying potential confounding covariates from ~90 time-varying and time-invariant sociodemographic variables and lagged exposures. We then fit MSMs for each adversity exposure and outcome combination to examine the putative causal influence of various patterns of each socioeconomic adversity exposure over infancy (6-15 months of age), toddlerhood (24-35 months of age), and early childhood (58 months of age) on maternal sensitivity and child positive affect at 58 months.

Preliminary results indicate the presence of linear and nonlinear dose effects (e.g., Figure 1) that differ by both the dimension of socioeconomic adversity exposure and outcome. Further, some results are suggestive of the importance of the timing that could reflect sensitive periods during which the effects of exposure to some dimensions of socioeconomic adversity may uniquely impact future parent-child interactions. Findings suggest that dimensions of socioeconomic adversity may differentially influence parent-child interactions as a function of both the dosage and developmental timing of exposure. This study offers both methodological and substantive advances to developmental science, harnessing a causally informed approach to help elucidate the processes through which different forms of socioeconomic disadvantage shape parent-child interactions in the first years of life.

Authors

• Isabella Chask'a Stallworthy, University of Minnesota - Twin Cities
  Presenting Author

• Meriah L. DeJoseph, University of Minnesota - Twin Cities
  Non-Presenting Author

• Clancy Blair, New York University
  Non-Presenting Author

• Daniel Berry, University of Minnesota - Twin Cities
  Non-Presenting Author