ABSTRACT
BACKGROUND: Social support is commonly examined as a protective factor for children with a history of child maltreatment, and it has been measured by self-report via the Social Support Scale for Children (SSSC). Although the SSSC has established adequate reliability and validity in community and clinical samples, its psychometric properties have yet to be assessed in a sample of foster care youth. OBJECTIVE: This study provided a psychometric comparison of the SSSC in youth residing in foster care with youth residing in the community. PARTICIPANTS AND SETTING: Participants were two, comparable samples of 214 youth participants residing in foster care and 163 youth participants from the community between the ages of 8-12 years. METHODS: Community participants were recruited from local middle schools, and an age-matched comparison sample from a larger study on youth in foster care was utilized for comparison. Youth self-reported across measures and provided demographic information. Confirmatory factor analysis was utilized to determine measurement model fit to the data, and invariance testing was conducted to compare measurement models across the samples. RESULTS: Differences between samples in the factor structure and item distribution of the SSSC emerged. Specifically, the community sample provided adequate fit to the original four-factor model (friend, classmate, teacher, parent) of the SSSC, whereas the foster sample required a three-factor model (combined friend and classmate constructs). The newly defined three-factor model provided significant associations with youth behavioral and emotional outcomes. CONCLUSIONS: Youth in foster care may perceive social support across sources differently from youth residing in the community.