Characterizing the Social Networks of Foster Youth Aging out of Care

Background and Purpose: Foster youth aging out of the child welfare system report adverse health outcomes and often struggle to demonstrate economic self-sufficiency, largely because of their limited human capital. These foster youth report that they have social supports they can rely on, which raises questions as to the disparity between the social supports they have and the supports required for successfully transitioning to adulthood. While mentoring programs may be able to provide the necessary social supports these youth lack, more information is needed to understand how these youths’ social networks should be expanded for greatest positive impact. Research on mentoring programs have been conceptually limited by a focus on the emotional support and socioemotional skill development that mentors provide (i.e., bonding capital) to the exclusion of bridging capital, which has been theorized as crucial to the success of young adults in today’s labor market. This pilot study aims to characterize the current social networks of older youth in foster care involved in a local mentoring program, and provides a comparison to the social networks of college students in a 4-year university.

Methods: Older foster care youth entering a local mentoring program over a one-year period (N = 15) were recruited for participation in this study. A convenience sample of college students from a 4-year university were also recruited from the social work department. Baseline interviews were conducted by trained interviewers over Zoom and recorded. Participants were asked to identify people who provide them with social support in four domains: family (chosen or biological), friends, school/work, and community. The participants were prompted to identify people in these four domains who provide various types of social support, including emotional, positive social interactions, informational, and concrete support. Ego networks for each of the study participants were analyzed using the Networkx package in Python 3.8.3.

Results: The foster youth in the mentoring program reported strong social support networks, and include biological family members. Their networks comprised of higher network density than the college students (0.78 vs. 0.44), indicating that there are more connections in their support networks than the college students. Additionally, the foster youth report higher clustering coefficients (and 0.73 vs. 0.23), indicating that their networks are more tightly connected while college students have with multiple cliques in their support networks. At the same time, the foster youth support networks also mentioned adults from the mentoring program, albeit as isolated nodes.

Conclusions and Implications: Foster youth aging out of care report having strongly interconnected social support networks. However, in comparison to college students, it appears that their networks lack the bridging capital that would link them to the social networks that could provide useful educational and employment opportunities. This suggests that adding “weak ties” to these youths’ networks through the mentoring program may be a promising approach. We hypothesize that, if those isolated nodes representing adults from the program grow into cliques in their social networks, the network will then provide the foster youth with invaluable opportunities necessary to successfully become independent.