Networks of Evidence Use in the Context of Federal Childhood Obesity Policymaking

Matthew Weber, Itzhak Yanovitzky


Despite considerable interest in strategies for promoting evidence-informed policymaking, research that tracks and analyzes the use of research evidence in public policymaking is scarce. We use social network analysis to trace the flow of research evidence in the context of childhood obesity legislation from 2000-2014. Network analysis is used to examine how political homophily, or policymakers’ shared group affiliations, and various structural characteristics of the policy network influence evidence exchange over this period. Current thinking about the use of research evidence in policymaking has evolved toward a broader perspective on the research-policy link that recognizes that research use unfolds within a social ecology of relationships, organizational settings, and political and policy contexts. Central questions are thus how, when, and why research evidence is acquired and used by decision-makers in the context of a policy process that is dynamic, political, and involves multiple social actors working across a number of distinct systems. In order to test these mechanisms, we focus on the context of childhood obesity, and specifically focus on federal policymaking efforts from 2000 to 2014. Childhood obesity has become one of the nation’s most serious problems affecting children and youth with several lifelong co-morbidities, including cardiovascular disease, type 2 diabetes, hypertension, osteoarthritis, cancer, and psychosocial burdens. To enable analysis of the aforementioned questions, data were extracted from content analyzing a large corpus of congressional documents. Our selection of relevant documents was designed to adequately represent the entire policymaking process, beginning with the introduction of bills, through committee hearings, and to floor debates leading to voting on proposals. We chose to analyze only documents directly relevant to childhood obesity prevention efforts and published between January 1, 2000 and December 31, 2014. We retained a total of 786 documents for further analysis (206 congressional bills, 190 committee hearings and reports, and 390 records of floor debates). The resulting evidence exchange network is comprised of 286 total actors and 129,222 edges, which was analyzed using exponential random graph models. Our study finds that the combined effects of homophily as well as structural connectedness increase the probability that policymakers share evidence in this domain. More specifically, we show that being connected to a specific range of actors in a policy network may create a space where individuals are more likely to share research evidence. Together, these findings help explain key aspects of evidence use to develop legislation and builds theory regarding how evidence is exchanged across policy domains. Further, our findings show that being connected to at least four to six other actors in a policy network may create a “structural sweet spot” where actors are more likely to exchange evidence from one actor to another. When policymakers served on one or more special committees, especially when affiliated with the core Senate committees debating childhood obesity legislation, the proclivity to share evidence became substantially more likely. Together, our findings provide important insights for how policymakers advance evidence-based policy in the domain of childhood obesity legislation.

← Schedule