Session: Applications of Social Network Analysis to Substance Use Prevention Program Development, Program Impact Evaluation and Diffusion Analysis (Society for Prevention Research 22nd Annual Meeting)

SESSION INTRODUCTION: The goal of this session is to demonstrate the potential of concepts and methods from social network analysis to enhance the impact of substance use prevention efforts in adolescence. The symposium includes papers from a single research group but addresses three distinct phases of prevention research: program content (development), program impact evaluation, and examination of diffusion of intervention effects.

The first paper, “Peer Processes Targeted within Universal School-Based Substance Use Prevention Programs,” reports the results of a content analysis of three leading school-based substance use prevention programs. It focuses on the extent to which key peer processes identified in basic developmental research are addressed by each program. Results indicate that programs vary in overall attention to peer processes, focus mainly on peer influence (rather than peer selection) and generally do not mention the possibility of positive peer influences.

The second paper, “Clarifying Intervention impacts on Social Networks: The roles of baseline levels of problem behavior and network centrality” examines two alternative processes that could account for recent findings that the PROSPER intervention alters peer networks in ways that reduce the potential for peer influence toward problem behavior. Preliminary results suggest that program impacts on peer influence may be partly explained by disproportionately strong impacts on the problem behavior of youth who were initially high in network centrality (influence potential).

The third paper, “Predicting the Diffusion of Intervention Effects Using Social Network Analytic (SNA) Measures”, uses data from PROSPER to demonstrate how measures derived from social network analysis can clarify how intervention effects can diffuse from youth who participated in the family-based intervention (~20% of sample) to their classmates (the other 80%). SNA measures capturing the location of intervention participants in the friendship network and the structural features of the friendship network itself accounted for the extent of diffusion observed 2 years post-intervention.

The discussant, a leading researcher on the use of social network methods in prevention research, will provide integrative comments and will moderate discussion among presenters and individuals attending the session. The application of social network concepts across a range of issues (intervention development, program impact evaluation, diffusion of effects) should attract interest from a diverse range of SPR researchers, including those who are familiar with network approaches and those who are beginning to explore their potential application.