An ecological view of executive function in young children: Variation in and predictors of executive function skills over one school year

For many years, researchers studied executive functions (EFs) in the laboratory with a focus on understanding an individual child's development and brain processes in a controlled environment. Building on this foundational research, there is a growing interest in EFs in the context of a child's dynamic, social world, and the contextual and compositional factors influencing EF development. This paper provides a descriptive view of EFs in 1112 K-3 children from six schools in Phoenix, AZ, USA. The study's goals were to examine (1) variation in EF scores between and within schools and classrooms, (2) predictors of variation in children's spring EF scores, and (3) individual and compositional predictors of children's spring EF scores. Our findings indicate greater variation in children's EF within schools than between, with very little or no variation arising from differences between schools. Though we observed greater variation within classrooms than between them, a notable amount of variance in children's spring EF scores appears to arise from differences between classrooms. Classroom-level variables, including a fall leave-out classroom mean (without the students' own score) and the number of children in the top or bottom grade-level quartiles in each classroom, were significant predictors of variation in spring EF scores as well as in fall to spring changes in EF. In some cases, the classroom variables were stronger predictors than individual fall scores. Findings suggest that understanding variation and cultivating growth in EF skills requires intervention, measurement, and analytic approaches that extend beyond the individual to include compositional features of the classroom environment. Research HighlightsVariation in children's EF scores (n = 1112 children) was greater within schools (n = 6 schools) than between, with very little or no variance arising from differences between schools.While variance was greater within classrooms than between (n = 67 classrooms), a notable amount of variance in children's spring EF scores appears to arise from differences between classrooms.Classroom-level variables (e.g., leave-out mean, number of children in the top or bottom grade-level quartiles in each classroom) were significant predictors of variation and of changes in spring EF outcomes.In some cases, the classroom variables were stronger predictors of spring EF than individual fall scores.