Session: Leveraging Social Network Analysis to Improve Prevention Programs (Society for Prevention Research 24th Annual Meeting)

SESSION INTRODUCTION: Recent work highlights the potential of network analysis to advance prevention science (e.g., Gest et al., 2011). This symposium demonstrates innovative applications of social network analysis (SNA) across three intervention contexts aimed at promoting youth well-being and reducing health risk behaviors within school settings. The three proposed papers demonstrate unique ways in which SNA can be used to better understand and improve prevention programs and their effects at two different stages: (1) processes that are active during program implementation, and (2) processes that account for the diffusion of program effects after implementation is complete.

The first paper uses SNA to measure group process within a group-based mentoring program with college women mentors and middle school girl mentees.  Analyses test whether SNA-based indices of participants’ connectedness within these groups help to account for between-person differences in the program benefits accrued to members. In the second paper, SNA is used to identify key opinion leaders among teachers within 17 elementary and middle schools.  Analyses examine teacher characteristics associated with being a key opinion leader in order to inform strategies to improve implementation fidelity and dissemination of an intervention developed to reduce the discipline gap in schools.  Finally, the third paper uses SNA to examine the extent to which effects of a peer-led suicide prevention program may diffuse over time to non-participants through friendship networks. Longitudinal friendship network data are used to track friend selection and influence processes with respect to suicidal ideation within high school friendship networks.

The three proposed abstracts highlight important ways in which SNA can be leveraged to enhance understanding of and ultimately improve prevention programs.  The first two abstracts focus on centrality scores: as indicators of participants’ connectedness as they relate to program outcomes in the first paper, and as a means of identifying key opinion leaders in the second paper.  The second and third abstract focus on diffusion of intervention effects through peer-nominated leaders: the first within a teacher network, the second within an adolescent network.  All three abstracts examine prevention programs that have the ultimate goal of improving well-being and reducing health risk behaviors. The application of SNA to analyzing data from these programs has great potential to shed new light on how such programs work to improve outcomes for participants and diffuse to non-participants.  The contributions of these papers will be enhanced by comments from a discussant who is a leader in the field of network effects on health and health risk behaviors.