Spanking and children's early academic skills: Strengthening causal estimates

Abstract

Using the Early Childhood Longitudinal Study – Kindergarten Class of 2010–2011, this study examined the associations between spanking experiences and children's subsequent reading and math scores. Two different groups were matched on a range of family sociodemographics and children's initial reading and math scores using entropy balancing: not spanked vs spanked (Ns=17,171–17,537); and not recently vs recently spanked (Ns=10,393–10,724). Sample sizes were provided in ranges because they varied across multiple imputed data sets. The results of the lagged dependent variable regression analyses showed that a lifetime experience of spanking by age 5 did not predict children's subsequent reading and math scores. However, children who were recently spanked at the age of 5 showed significantly lower reading and math scores at ages 6 and 7 compared with those who were spanked but not recently. These results, replicated using different estimation methods (i.e., difference score analyses), strengthen the argument that spanking may impair early reading and math skills.