The Battle is On: Factors that Motivate People to Combat Anti-Vaccine Misinformation

ABSTRACT

This study proposes a theory-driven model to concurrently examine the cognitive and emotional factors that motivate vaccine supporters to combat erroneous online anti-vaccination information. The model was tested using data from a web survey of 599 vaccination supporters in the United States. The vaccine supporters reported greater support for government regulation of misinformation when they perceived greater susceptibility among the general public to the influence of misinformation. Surprisingly, the perceived severity of the influence was inversely related to respondents’ intention to correct misinformation. In addition, perceived susceptibility to the influence of anti-vaccine misinformation and perceived severity of its influence on others induced negative emotions that included anticipated guilt and anger. The negative emotions in turn motivate vaccine supporters to attitudinally support government’s media restriction or behaviorally correct the online misinformation.

Notes

1. The eight items of information were (1) “Vaccines contain poisons, toxins, or contaminants” (M = 2.53, SD = 1.04); (2) “Getting multiple vaccines at the same time causes chronic health problems” (M = 2.28, SD = 1.10); (3) “Alternative medicines such as homeopathy, vitamins, or essential oils can replace vaccination” (M = 2.37, SD = 1.14); (4) “Vaccine-preventable diseases are uncommon, not contagious, or relatively mild” (M = 2.30, SD = 1.15); (5) “Vaccines are not effective in general” (M = 2.35, SD = 1.05); (6) “Vaccines cause serious complications or have serious side effects” (M = 2.78, SD = 1.02); (7) “Vaccines cause illnesses of unknown origin, such as autism or sudden infant death syndrome (SIDS)” (M = 2.70, SD = 1.11); and (8) “People die after taking vaccines” (M = 2.14, SD = 1.10).

2. The full graph for the confirmatory factor analysis can be obtained from the corresponding author upon request.

3. Originally, we used four items to measure anticipated guilt. However, the reversed item related to feeling sorry (“I would not feel sorry if I did not try to correct the misinformation when I saw it online”) was removed because of its poor factor loading (−0.23).

4. Originally, we used six items to measure corrective behavior. However, three items (i.e., reporting the post so that it gets removed, unfollowing or blocking the person who posted the wrong information, and ignoring the wrong post) were removed because of their low factor loadings (0.68, 0.58, and 0.16, respectively; all < 0.70).

5. The information about the associations between control variables and pro-vaccination people’s support for regulating anti-vaccination misinformation and their intention to correct anti-vaccination misinformation is available from the corresponding author.

6. Some have argued that to examine the significance of indirect effects and to avoid potential statistical problems arising from the asymmetric and nonnormal distributions often associated with indirect effects, researchers should take account of the distribution of the product to identify more exact confidence limits than those based on normal distribution (MacKinnon et al., 2007). This approach ensures higher testing power and lower type I error rates than other commonly used methods (MacKinnon et al., 2007).

Additional information

Funding

This study is supported by a grant from the City University of Hong Kong (SRG7004935) and the Public Policy Research Funding Scheme (2020.A1.093.20A) from the Policy Innovation and Co-ordination Office of the Government of the HKSAR awarded to the second author.