This project will use methods from quantitative anthropology to describe the social space of vaccine beliefs that circulate among the general public and to provide an initial assessment of how different belief variations influence decisions to vaccinate. The results will establish, for the first time, the patterns of co-variation in the wide variety of pro- and anti- vaccine beliefs, and which axes of this variation appear associated with decisions to vaccinate. Vaccination is a key public health defense against infectious disease, but the lay public largely does not fully appreciate scientific evidence when making decisions for or against vaccination. Understanding the inter-correlations of these beliefs, therefore, is imperative for designing effective educational interventions that can directly interface with the cultural beliefs that surround vaccination and influence the public's decision making on this issue. The project will leverage insights from two very different but complementary data sources: responses to a nationally representative survey (fielded on the RAND American Life Panel) and social media data from Twitter. Our analytic approach will begin with systematic coding techniques from mixed-methods research to classify vaccine beliefs into a comprehensive set of belief variants. Manual coding will be validated through inter-observer reliability checks and replicated at scale with machine-learning algorithms. Having systematically coded the data, we will then assess whether nationally representative survey data and data mined from Twitter produce similar results using Cultural Consensus Analysis, a technique from quantitative cultural anthropology. From the survey data we will test whether vaccine beliefs are correlated with decisions to vaccinate after controlling for demographic attributes. To ensure completion of this innovative and methodologically expansive project, the project team combines expertise from anthropology, decision science, clinical medicine, and biomathematics. The principal investigator brings to this project multiple years of both academic and industry experience in statistical modelling of cultural data.
Vaccination is a key public health defense against infectious disease, but patients' vaccination decisions may be more influenced by broadly circulated cultural beliefs than they are influenced by scientific evidence. This proposed research will systematically map the diversity of the publics' vaccination beliefs, assess how these beliefs influence vaccination decisions, and advise policy makers how to interface more directly with these popular belief systems that are critical to effective vaccination efforts.
