The Politicization of Health and Science: Role of Political Cues in Shaping the Beliefs of the Vaccine-Autism Link

ABSTRACT

One critical lesson learned from public opinion research about climate change is that the cost of politicization is disastrous. Although the literature has shown the dire consequences of politicized science issues, few have examined how such politicization is possibly triggered by political leaders in a seemingly nonpartisan science topic. Using two experiments (total n = 1,249), this article demonstrates how political cues over scientific expertise shape individuals’ beliefs in the vaccine and autism debate. The results indicate that Republicans tend to follow President Trump compared to scientists in the subject matter. On the other hand, Democrats follow scientists but are not influenced by Trump. The implications of political encroachment into health and science are discussed.

Notes

1. As Study 1 focuses on vaccine risks regarding MMR vaccines and autism, we conducted additional analyses only with participants who reported they had children (n = 484). In a nutshell, the analysis yielded identical results with the results with all participants. Specifically, in a mixed analysis of variance (ANOVA) with a 2 (source cues: political cues vs science cues) X 2 (vaccine risk: high vs. low) X 3 (party membership: Republicans, Independents, and Democrats), the 3-way interaction term was significant, F(1,484) = 6.07, p <.01, partial η² =.025. This indicated that the effects of political and scientific cues were significant but differed across party membership. Like the reported results with all participants, political cues did not influence Democrats, t(73) = 0.09, p =.93 and Independents, t(68) = 0.16, p =.87, but influenced Republicans, t(83) = 3.65, p <.001. Additionally, scientists influenced Democrats, t(94) = 4.01, p = <.001 and Independents, t(84) = 2.45, p =.02, but not Republicans, t(70) = 0.56, p =.58. The analysis with participants without children also showed seemingly similar patterns but the 3-way interaction term was not significant mostly due to the small sample size (n = 164).

2. In addition to previous literature review, we confirmed this by testing the baseline vaccine belief among partisans who were assigned to our control condition in Study 2.