Analyses are based on two waves of longitudinal data from the National Survey of Children's Exposure to Violence (NatSCEV), conducted in 2008 and 2010, that comprised a telephone sample of 1,186 youth ages 10 to 17. Using structural equation modeling, the study examined direct and indirect effects on distress symptoms of increased, decreased, and stable high poly-victimization between Waves 1 and 2 compared to no or low victimization in both waves. Specifically, the study considered the extent to which reductions in core psychosocial resources - including family support, peer support, self-esteem, and mastery - mediate the relationship between these poly-victimization conditions and distress. Relative to stable low victimization, both increased poly-victimization and stable high poly-victimization were associated with declines in all four resources; however, only self-esteem and mastery significantly mediated the association between poly-victimization and distress, with mastery showing the strongest effect. Although significant indirect effects were evident, poly-victimization still had a strong direct effect on distress with resource factors controlled. Findings support the hypothesis that the potent effect of poly-victimization on youth mental health is, in part, due to its damaging influence on core psychosocial resources. (Publisher abstract modified)