Multiple informants and developmental timing: Predicting adulthood outcomes from childhood internalizing problem behaviors

• In Event: Substantive Developmental Research Questions Addressed with Applied Measurement Methods

Abstract

The aim of this study was to examine if and how youth internalizing problem behaviors across childhood and adolescence are connected to consequential life outcomes in adulthood and if these trends vary across different informants of child behavior. That is, different informants of children’s internalizing behaviors (mothers, fathers, teachers, and children themselves) tend to only weakly converge, which raises questions about if certain informants provide stronger reports for predicting long-term outcomes. Using data from a longitudinal, high-risk sample (N=1069) that assessed children every three years from early childhood through late adolescence (age 3-17) and again in adulthood (aged 24-31), the present study examined these questions using two methods. First, latent growth models were fit for each informant separately to the repeated assessment of internalizing problems across childhood (maternal and paternal report spanning children aged 3-17; teacher report spanning age 6-17; and children’s self-report spanning age 11-17), and the latent growth factors were used to predict the adult outcomes, providing a holistic sense of how internalizing problems across childhood and adolescence were related to these outcomes. Second, a series of trifactor models were fit at different ages (early childhood, middle childhood, adolescence) to both determine the relative importance of internalizing problems during particular developmental periods for later outcomes as well as to integrate multiple informants' reports. In these models, both the consensus and informant-specific factors were used to predict later outcomes. In the latent growth models, mother- and teacher-reported internalizing problems were predictive of several adult outcomes, but father- and child-reported problems were not; teacher reports of internalizing problems were particularly robust predictors of adult income, educational attainment, and partner-reported romantic relationship quality. The trifactor models suggested that internalizing problems during mid- and late-adolescence were particularly predictive of later outcomes compared to internalizing problems during early and middle childhood. Notably, informant-specific factors, most often the teacher-specific factor, frequently contributed to outcome prediction above and beyond the consensus factor. Implications of these results for the selection of informants in research and clinical settings, with particular emphasis on the value of teacher reports, will be discussed.

Authors

• Allison Gornik, Kennedy Krieger Institute
  Presenting Author

• C. Emily Durbin, Michigan State University
  Non-Presenting Author

• Robert A Zucker, University of Michigan
  Non-Presenting Author