Using experimental variation to examine the (co-)development of cognitive and social-emotional skills in early childhood

How cognitive and social-emotional skills emerges in childhood is a central question in developmental psychology. Despite a proliferation of theory and empirical work, the field has struggled to establish causal effects. We employed a novel approach to test the stability and codevelopment of cognitive and social-emotional skills in early childhood using longitudinal randomized controlled trial data from the nationally representative Head Start Impact Study (n = 4,667). Capitalizing on the study’s clustered design, we computed treatment effects on both skills for each cluster (k = 84). Using meta-analytic techniques, we found that changes to children’s cognitive skills persisted at a rate of approximately 40% one year after program end and 30% two years after program end. Changes to social-emotional skills persisted at a rate of approximately 20% at both timepoints, though estimates were statistically non-significant. We did not observe strong evidence of transfer effects from cognitive to social-emotional skills or vice versa, as estimates were generally imprecise and statistically non-significant. However, future work should investigate cognitive to social-emotional transfer processes given that, above and beyond the initial impact of Head Start on social-emotional skills, a 1 SD increase in cognitive skill impacts consistently predicted a ~ .20 SD (non-significant) increase in impacts on social-emotional skills one and two years later. A comparison of estimates from our new method and traditional correlational methods indicated that traditional methods produce upwardly biased auto-regressive, but not cross-lagged, estimates. We encourage researchers to use within-study meta-analysis to test psychological theories.