Over the life of the project, we have conducted tens of thousands of survey interviews to better understand public attitudes and behavior on these key topics. As a team, we have published over 30 scientific peer reviewed journal articles in some of the highest-ranked scientific journals, such as Proceedings of the National Academy of Science (PNAS) and Nature Human Behaviour. The project team has similarly used multiple avenues for disseminating results as widely as possible. In addition to scientific journal articles, we have presented work to academic and scientific audiences at dozens of conferences, workshops, and seminars; we have presented lectures and keynote addresses to public audiences; we have issued press releases about our work and talked to reporters (who have helped bring our results to the general public); and we have participated in television and radio programs to speak about project result directly with the public.
Importantly, the project has amassed a significant body of work that shows that corrections are effective at reducing misperceptions. This research is an important and effective counter-point to earlier research (by the PI) that giving people information to correct misperceptions would “backfire”, causing people to believe incorrect facts even more strongly. However, there are some important and notable caveats about the effectiveness of corrections. First, the effects of corrections go away over time, and fairly quickly – this dissipating is a process of days or weeks rather than months or years. Second, changing factual beliefs may only have limited effect on related beliefs, such as how people perceive politicians or they support for different policies.
The project has also amassed a large body of work concerning “fake news.” In a landmark study of the consumption of this type of content during the 2016 US Presidential election, we discovered aspects that alarming as well as aspects that are reassuring. On the alarming side, we find that a very high percentage of people had visited at least one “fake news” website. On the reassuring side, we also found that “fake news” comprises only a very small share of the news-related websites people visit online. We find very strong evidence for a selective exposure explanation, where people are visiting content that matches their ideological preferences. We also show that social media played a key role in how people were exposed to “fake news.”
We were also successful in identifying one way to combat the spread of “fake news” through a digital media literacy intervention. We found that a digital media literacy intervention increased people’s ability to better discern between mainstream “real” news and untrustworthy “fake” news. We also have evidence that the effect of this intervention persists, though the effect decreases over time. There is one important caveat. Our intervention increased skepticism in both mainstream and untrustworthy news, but did so much more “fake news.” While the intervention works, future research should focus on how to increase discernment without also increasing skepticism towards all types of content.
On the topic of vaccination, we focused on what factors affected people’s willingness to get a COVID-19 vaccine. We found that the predicted efficacy and the likelihood of side-effects were particularly important considerations. We also found that some factors mattered very little – such as the underlying vaccine technology (e.g. mRNA). In terms of understanding vaccine attitudes and behaviors at a societal level, we were able to provide good evidence that aggregated survey data at the regional level in Europe that measure vaccine attitudes can effectively explain vaccination uptake, validating an important tool for public health officials in responding to public health threats.
