Children at Risk of Academic Failure: How Child Health and Social-Emotional Skills Affect Reading and Mathematics Achievement from Kindergarten through Fifth Grade.

Abstract

One of the most enduring problems facing educators in America is that a large number of children underachieve academically relative to their peers upon school entry and continue to underachieve throughout their schooling. It is widely assumed that knowing what child factors cause academic success or failure can eventually lead to the development of interventions that target those child factors and improve the academic achievement of children at risk. This dissertation explores how children’s health and social-emotional skills affect their reading and math achievement trajectories over the elementary school years. Specifically it investigates which constructs within these two domains have the strongest effect on reading and math achievement, and the size of these effect over time, when accounting for repeated measures of health and social-emotional skills. Hierarchical cross-classified models (HCM) of reading and math achievement from kindergarten through fifth grade are created to explore these issues, using data from the Early Childhood Longitudinal Study-Kindergarten Class of 1998-1999 (ECLS-K). The cross-classified structure of these models accounts for the different schools ECLS-K children moved through over the elementary school years. Repeated measures of health and social-emotional skills are included in the HCM models either as time-varying covariates or as stable traits. The study finds that many of the child health and social-emotional skills measures included have statistically significant effects on reading and math achievement at kindergarten entry and throughout elementary school, holding all other measures in the model constant. The measure of children’s regulatory behaviors (a factor based on the average of four highly correlated scales from the teacher social rating scale) has the strongest independent effect on achievement of all health and social-emotional skills measures, as well as of most of the background control measures included in the model. The standardized effect sizes of the regulatory behaviors factor on reading and math achievement at kindergarten entry is 0.18 and 0.23, respectively, and increases to 0.26 and 0.29, respectively, by fifth grade. The regulatory behaviors factor accounts for almost half of the total achievement gap (calculated using all covariates in the model) between children with poor regulatory behaviors and the average child.