Peer Victimization and Prospective Changes in Children's Inferential Styles

Abstract

There is evidence that the cognitive vulnerabilities featured in the hopelessness theory of depression—inferential styles for the causes, consequences, and self-worth implications of negative events—increase risk for depression. Given this, it is important to understand how these inferential styles develop. In this study, we examined the impact of overt and relational peer victimization in a multiwave prospective study of 100 children (8–12 years of age) with peer victimization and inferential styles assessed every 2 months for 6 months (4 assessments total). Overt victimization uniquely predicted prospective changes in children's inferential styles for consequences and relational victimization uniquely predicted changes in inferential styles for self-characteristics. It is important to note that these relations were maintained even after controlling for the impact of concurrent depressive symptoms. These results add to a growing body of research suggesting that peer victimization may increase risk for the development of cognitive vulnerability to depression in children.

Acknowledgments

This project was supported by National Institute of Child Health and Human Development grants HD048664 and HD057066 and by funding from the Center for Development and Behavioral Neuroscience, Binghamton University, awarded to B. E. Gibb. We thank Jessica Benas, Dorothy Uhrlass, and Marie Grassia for their help in conducting assessments for this project.

Notes

Note: Mother MDD = mother history of major depressive disorder (yes = 1, no = 0); child gender (girl = 1; 0 = boy); SEQ-OV = Social Experiences Questionnaire–Overt Victimization subscale; SEQ-RV = Social Experiences Questionnaire–Relational Victimization subscale; CASQ = Children's Attributional Style Questionnaire; Cons = Children's Cognitive Style Questionnaire–Consequences; Self = Children's Cognitive Style Questionnaire–Self-Characteristics; CDI = Children's Depression Inventory.

Correlations ≥ .20 significant at p < .05, correlations ≥ .26 significant at p < .01, and correlations ≥ .33 significant at p < .001.

Note: The DV in this analysis is Children's Inferential Style for Self-Characteristics at Time T. Mother MDD = mother history of major depressive disorder (yes = 1, no = 0); SEQ-RV = Social Experiences Questionnaire–Relational Victimization subscale; SEQ-OV = Social Experiences Questionnaire–Overt Victimization subscale; Self_T-1 = Children's Cognitive Style Questionnaire–Self-Characteristics at Time T-1; CDI = Children's Depression Inventory.

¹For interested readers, we should note that the Child Gender × Relational Victimization interaction was significant at the traditional .05 level predicting changes in children's inferential styles for consequences, t(147.81) = 2.05, p = .04, r _{effect size}  = .17. Examining the impact of relational victimization separately in boys and girls, we found that it predicted changes in inferential styles for consequences among boys, t(58.20) = 3.10, p = .003, r _{effect size}  = .38, but not girls, t(99.30) = −0.32, p = .75, r _{effect size}  = −.03. Among boys, this effect was maintained when statistically controlling for concurrent levels of overt victimization and depressive symptoms, t(64.91) = 1.98, p = .05, r _{effect size}  = .24. In addition, the Child Age × Relational Victimization interaction was significant at p < .05 in predicting changes in children's inferential styles for causes, t(179.04) = 2.04, p = .04, r _{effect size}  = .15. Examining the impact of relational victimization separately in younger (8–9-year-olds) versus older (10–12-year-olds) children, we found that it predicted changes in inferential styles for causes among older, t(111.57) = 2.04, p = .03, r _{effect size}  = .19, but not younger, t(74.16) = −0.76, p = .45, r _{effect size}  = −.09, children. However, this effect was reduced to nonsignificant when we statistically controlled for concurrent levels of overt victimization, t(131.13) = 1.36, p = .18, r _{effect size}  = .12, or depressive symptoms, t(142.27) = 0.41, p = .68, r _{effect size}  = .03.