A computational approach to study the gap and barriers between science and policy

Abstract Research on the gap between science and policymaking and the barriers to utilizing science in policy remains controversial, since previous research methods have dominated data collection with perceived participant responses. This approach makes research findings applicable to particular contexts, particular participants, and particular times but cannot be generalized. We used a computational model to analyse linguistic text data from two communities and compared the results with self-reported research findings. The outcomes support that scientists and policymakers have their own linguistic characteristics, with scientists in the medical field preferring jargon and policymakers’ language containing public satisfaction pressures. Language representation also validates the existence of barriers in the use of science