Integrating Social Network Data into Community Assessments

Background and Purpose:Asset mapping provides a strength-based approach to community assessment. Communities conducting asset mapping typically identify existing resources (often programs) and areas for development. This approach provides foundational information but more detailed program data may lead to a richer understanding of the community.

Relationships among programs provide additional information about each program as well as the community as a whole. Network analysis measures and visualizes relationships among actors, in this case, programs, and therefore concretely describes programs’ connectivity and the community’s service-delivery environment. We therefore added network questions to asset mapping of programs whose goals were aligned with those of a community coalition engaged in the collective impact process. Because the coalition established a goal of improving college and career attainment in their community, we bounded the network as programs who contributed to this goal. We assessed collaboration and referral networks among these programs with the goal of contributing to community assessment and planning. Additionally, our focus on programs provides a different perspective than previous network studies of coalition leadership.

Methods: This study was set in a small, Midwestern city with a population of less than 50,000. We conducted phone interviews in 2016 with primary contact persons for programs whose goals aligned with those established through the coalition’s collective impact process. Respondents were primarily identified by the community coalition, or secondarily, through the course of interviews. Using open-ended questions, we asked respondents to identify programs with which they collaborated, to which they referred, and from which they received referrals. UCINet was used to visualize and analyze network data.

Results: Programs (N=96) provided educational, basic needs, and youth development programming. Programs served 1-50,000 community members annually. Most (55%) programs served high school aged youth.

Programs reported few collaborations (Mean=1.4). In-degree directed ties for collaboration (ties respondents reported) ranged from 0 to 11 (Median=1). Out-degree directed ties (ties others reported about the respondent’s program) for collaboration were concentrated in a few programs (Median=0; Range of 0 to 25). Betweenness, a measure of network centrality, indicates programs that connect otherwise unconnected clusters. Grant-making, arts, community college, and afterschool programs had the highest collaboration betweenness.

Referral networks were similarly sparse (Mean=1.2). In-degree directed referrals ranged from 0 to 7 (Median=0) and out-degree referrals (ties others reported to program) ranged from 0 to 11 (Median=0). Respondents most commonly referred program participants to basic needs (n=5), children’s activities fund, family case management, and recreation programming. Programs with high referral betweenness included many of the same programs. Educational programs—the focus of the community’s intervention—reported fewer referrals and were less central to those networks.

Conclusion and Implications: Results from this study underlie the importance of assessing networks among programs when undertaking community-wide efforts, especially the collective impact process. In this community, increasing referral and collaboration networks has the potential to achieve equal opportunity for community members. Communities engaged in collective impact should consider incorporating network data in formative and summative evaluation.