The current study investigated concurrent and longitudinal mediated and mediated moderation pathways among maltreatment, self-perception (i.e., loneliness and self-esteem), social support, and internalizing and externalizing behavior problems. For both genders, early childhood maltreatment (i.e., ages 0–6) was related directly to internalizing and externalizing behavior problems at age 6, and later maltreatment (i.e., ages 6–8) was directly related to internalizing and externalizing behavior problems at age 8. Results of concurrent mediation and mediated moderation indicated that early maltreatment was significantly related to internalizing and externalizing behavior problems at age 6 indirectly both through age 6 loneliness and self-esteem for boys and through age 6 loneliness for girls. Significant moderation of the pathway from early maltreatment to self-esteem, and for boys, significant mediated moderation to emotional and behavioral problems were found, such that the mediated effect through self-esteem varied across levels of social support, though in an unexpected direction. No significant longitudinal mediation or mediated moderation was found, however, between the age 6 mediators and moderator and internalizing or externalizing problems at age 8. The roles of the hypothesized mediating and moderating mechanisms are discussed, with implications for designing intervention and prevention programs.
c1 Address correspondence and reprint requests to: Karen Appleyard, Center for Child and Family Health, 411 West Chapel Hill Street, Suite 908, Duke University, Durham, NC 27701; E-mail: email@example.com.
Portions of this work were completed while the first author was a postdoctoral fellow at the Center for Developmental Science at the University of North Carolina. An earlier version of this study was presented at the biennial meeting of the Society for Research on Child Development, April 2007, in Boston. The research was facilitated by grants to the third author from the National Center on Child Abuse and Neglect (NCCAN) and the Children's Bureau, Office of Child Abuse and Neglect (OCAN), US DHHS (90CA1746), by a grant to the first author from NICHD (Center Training Grant 5 T32 HD07376), and by the first author's participation in the National Data Archive on Child Abuse and Neglect (NDACAN) Summer Institute. The authors thank John Eckenrode and the NDACAN staff for their support and for their diligence in maintaining the archive. We are grateful to Antonio Morgan-Lopez of RTI International for his skilled and generous technical assistance with the mediated moderation statistical analyses, and to three anonymous reviewers for their thorough and thoughtful comments on the paper. The authors also are indebted to the children and families in the LONGSCAN Study for their long-term participation in the project and their contributions to our understanding of children's development.