Between expertise and worldview: The belief systems of scientists in Swiss COVID-19 policy debates

During the COVID-19 pandemic, scientists were important actors in the policy-making process in Switzerland. They not only participated in the policy process in formal advisory bodies, but also gave their assessment of the situation in a more informal way in public debates. However, the advice of the scientists was not always followed, and some scientists publicly expressed their disagreement with the measures taken to control the virus. This led to some controversy about the role of science in the management of the pandemic. In addition, there is often debate about whether scientists can be neutral advisors and the extent to which they are influenced by their own values and cultural backgrounds. This raises the question of how scientists interpreted the situation during the COVID-19 pandemic compared to other actors, and what values and beliefs played a role. According to the Narrative Policy Framework (NPF), assessing the belief systems behind policy narratives helps us understand how people give meaning to a policy process or policy object. The classification of belief systems in NPF is often based on Cultural Theory. Based on two dimensions - grid and group - the theory distinguishes between four cultural worldviews: Hierarchists, individualists, egalitarians, and fatalists. Building on these theories, I will examine the belief systems underlying scientists' narratives during the COVID-19 pandemic, how they changed over the course of the pandemic, and the extent to which they overlap with the belief systems reflected in politicians’ and governmental narratives. I will use supervised machine learning to classify the belief systems that Swiss scientists and other actors expressed in public debates about the COVID-19 pandemic based on the typology proposed in Cultural Theory. Therefore, this paper will not only give insights into scientists’ belief systems, but also contribute to a better understanding of how Natural Language Processing can be a tool in NPF research. I would like to propose the paper to the panel ‘Textual Data and NLP in Policy Process Research’.