Controlling contamination in child maltreatment research: Impact on effect size estimates for child behavior problems measured throughout childhood and adolescence

Abstract

Contamination, when members of a comparison or control condition are exposed to the event or intervention under scientific investigation, is a methodological phenomenon that downwardly biases the magnitude of effect size estimates. This study tested a novel approach for controlling contamination in observational child maltreatment research. Data from The Longitudinal Studies of Child Abuse and Neglect (LONGSCAN; N = 1354) were obtained to estimate the risk of confirmed child maltreatment on trajectories of internalizing and externalizing behaviors before and after controlling contamination. Baseline models, where contamination was uncontrolled, demonstrated a risk for greater internalizing (b = .29, p < .001, d = .40) and externalizing (b = .14, p = .040, d = .19) behavior trajectories. Final models, where contamination was controlled by separating the comparison condition into subgroups that did or did not self-report maltreatment, also demonstrated risks for greater internalizing (b = .37, p < .001, d = .51) and externalizing (b = .22, p = .028, d = .29) behavior trajectories. However, effect size estimates in final models were 27.5%-52.6% larger compared to baseline models. Controlling contamination in child maltreatment research can strengthen effect size estimates for child behavior problems, aiding future child maltreatment research design and analysis.

Keywords: child maltreatment; contamination; externalizing behaviors; internalizing behaviors.

Figure 1.

Effect size change in trajectories of internalizing behaviors before and after controlling contamination. Effect sizes were determined using the child maltreatment*age interaction term from Baseline and Final models and represent the differences in average slope trajectories between the respective confirmed child maltreatment and comparison conditions by the end of the study period. The No Confirmed Maltreatment comparison condition (Top) was separated into two subgroups (Bottom): 1) Contamination, defined as self-reported but no confirmed maltreatment, and 2) No contamination, defined as no self-reported or confirmed maltreatment.

Figure 2.

Effect size change in trajectories of externalizing behaviors before and after controlling contamination. Effect sizes were determined using the child maltreatment*age interaction term from Baseline and Final models and represent the differences in average slope trajectories between the respective confirmed child maltreatment and comparison conditions by the end of the study period. The No Confirmed Maltreatment comparison condition (Top) was separated into two subgroups (Bottom): 1) Contamination, defined as self-reported but no confirmed maltreatment, and 2) No contamination, defined as no self-reported or confirmed maltreatment.