Spanking, corporal punishment and negative long-term outcomes: A meta-analytic review of longitudinal studies

摘要

Social scientists continue to debate the impact of spanking and corporal punishment (CP) on negative child outcomes including externalizing and internalizing behavior problems and cognitive performance. Previous meta-analytic reviews have mixed long- and short-term studies and relied on bivariate r, which may inflate effect sizes. The current meta-analysis focused on longitudinal studies, and compared effects using bivariate r and better controlled partial r coefficients controlling for time-1 outcome variables. Consistent with previous findings, results based on bivariate r found small but non-trivial long-term relationships between spanking/CP use and negative outcomes. Spanking and CP correlated .14 and .18 respectively with externalizing problems, .12 and .21 with internalizing problems and - .09 and - .18 with cognitive performance. However, when better controlled partial r coefficients (pr) were examined, results were statistically significant but trivial (at or below pr=10) for externalizing (.07 for spanking, .08 for CP) and internalizing behaviors (.10 for spanking, insufficient studies for CP) and near the threshold of trivial for cognitive performance (- .11 for CP, insufficient studies for spanking). It is concluded that the impact of spanking and CP on the negative outcomes evaluated here (externalizing, internalizing behaviors and low cognitive performance) are minimal. It is advised that psychologists take a more nuanced approach in discussing the effects of spanking/CP with the general public, consistent with the size as well as the significance of their longitudinal associations with adverse outcomes.