Despite widespread recognition of the importance of behavioral health -- in its own right and in support of physical health -- and of the need to harmonize datasets -- including with rigorous psychometric procedures for measure linking -- most large public-use U.S. datasets -- including those funded by NICHD -- contain behavioral health assessments that have yet to be systematically and fully harmonized. Remedying this state of affairs provides an opportunity to rapidly increase the return on investment in these datasets and to deepen and broaden understanding about the ecological, economic, and social factors that support children's adaptive behavioral development, especially among socially disadvantaged children exposed to high-risk settings. The purpose of this project, therefore, is to harmonize a behavioral health measure (the Behavior Problems Index [BPI]) in two NICHD-funded datasets (the Children of the National Longitudinal Survey of Youth 1979 [C-NLSY79] and the Panel Study of Income Dynamics, Child Development Supplements [PSID-CDS]), and, in so doing, to examine whether the BPI works in equivalent ways between datasets and among demographic subgroups as well as to compare multiple strategies for linking to assure that scores are on a common metric. As a result, future studies can be sure that results reflect true subgroup differences in behavioral health, and not differences in the way mothers report about their sons and daughters or their younger versus older children or in the way that mothers vary in interpreting youths' behaviors across culture and context. We will demonstrate the utility of our approach by estimating differences in youths' behavioral health by child age, gender, and race-ethnicity; maternal education and family income; and region and urbanicity of residence. By archiving and disseminating the scores and code, the harmonized datasets will be readily accessible to other scholars who can use them to replicate and extend our work to numerous other important questions about how youths' behavioral health develops over time and what ecological factors and local resources support healthier development. To facilitate such future studies, we will also provide a crosswalk of variables relevant to a range of high-priority research questions that take advantage of the studies' multigenerational designs. In so doing, our work can stimulate existing users of the C-NLSY79 and PSID-CDS to use the datasets in new ways and for new users to work with the data. Interest in the datasets is demonstrated by the hundreds of peer-reviewed publications that have already examined the BPI in both datasets. Yet, these studies have had to rely on assuming that the BPI works in equivalent ways across subgroups, despite a growing body of studies demonstrating non-equivalence and despite scholars using ad hoc approaches or raising concerns about such assumptions. The harmonized datasets produced by the project will facilitate such work, offering greater power to examine important rare subgroups, opportunities for cross-study replication, and new insights into malleable factors that support children's healthy development.
This project will expand two public-use datasets in innovative ways by placing measures of youths' behavioral health on a common metric both between datasets and across subgroups within datasets, by harmonizing the subgroup variables as well as key design variables, by demonstrating the utility of the linked scores in comparison to the original raw scores, and by creating a crosswalk of variables needed for high priority research questions using the linked scores. Our results will improve estimates of behavioral health disparities among children and adolescents -- thereby suggesting priorities for prevention and intervention that can cascade into improved health and well-being for the population -- and the archived dataset will facilitate research on the behavioral health of young people consistent with the goals of Healthy People 2020 and the research priorities of NICHD. As such, this project is well aligned with the mission of NIH.
