Relational aggression and hostile attribution biases: testing multiple statistical methods and models

The present study used both categorical and dimensional approaches to test the association between relational and physical aggression and hostile intent attributions for both relational and instrumental provocation situations using the National Institute of Child Health and Human Development longitudinal Study of Early Child Care and Youth Development (M = 8.3 years-old, SD = 0.19, N = 840). A one standard deviation threshold was used to create categorical aggression status groups. In categorical analyses, children who were both relationally and physically aggressive were found to attribute more hostile intent attributions in relationally provocative situations. Results from the dimensional analyses indicated that female gender, physical aggression, and hostile attribution biases for instrumental provocations each predicted increases in relational aggression over time. Further, relational aggression was not predictive of hostile attribution biases for relational provocations, and instead physical aggression was predictive. Implications for the fields of developmental psychology and psychopathology are discussed.