865   Linked papers

What Makes People Attend to and Trust Science and Scientific Experts in Online Contexts?

In current controversial public debates (e.g., about climate change, vaccination), people discuss how the future of our societies and our relationship with technology should be shaped, and many of these debates proceed in online environments. While some people refer to scientific evidence and arguments, and also demand that science–as a system generating relevant knowledge for solving today’s problems–should be trusted, others neglect scientific arguments and evidence, or actively voice their distrust in science and technological developments. In this session, we raise the question of what makes people attend to and trust scientific information and scientific experts in online contexts. 

The first three papers investigate important factors of how individuals seek online information and which experts they perceive as trustworthy. Landrum et al. investigate the influence of gender and science curiosity on accessing scientific content on YouTube; Sharon et al. focus on the influence of an information seeker’s personal stance toward a topic for trustworthiness ratings of experts during information seeking in forums; and Yeo finds that not only a humorous presentation, but also source expertise explains how much people perceive comedy a valid source of scientific information. The final two presentations more broadly investigate factors that might benefit individuals’ acceptance of scientific information and trust in science: Flemming et al. introduce refutation texts as a means to enhance the acceptance of uncertainty in scientific communication, while Taddicken et al. investigate with a representative survey how individuals’ trust, knowledge and online use contribute to their problem awareness of climate change.

Author: Friederike Hendriks
University of Münster, Germany

The Role of Gender and Science Curiosity in Watching Science on YouTube

Speaker: Asheley R. Landrum
Texas Tech University, United States

Asheley R. Landrum, Dan M. Kahan, Dan Chapman, Othello Richards, Kristina Janet, Matthew Motta, Sevda Eris, Craig Rosa, Gabriela Quirós, Sue Ellen McCann. A number of professional media organizations create educational science videos for platforms such as YouTube. Though they make these videos for everyone who is curious about science, the videos struggle to gain female viewership. For example, public media company KQED Science produces a short science web series called Deep Look. Averaging across episodes, only 16% of Deep Look’s viewership is female. Such disparities lead science communication researchers and media professionals to ask whether the content they are producing—and/or the way they are producing it—is inherently off-putting to certain groups. Alternatively, it is possible that such disparities are not attributable to the content at all and are simply due to the ways in which YouTube’s algorithm suggests content. Collaborating with KQED, we examined this question with a large-scale survey experiment. Perhaps unfortunately, we replicated the gender disparity, suggesting that it is not simply a consequence of YouTube’s algorithm. We also discovered that the disparity occurs in the decision to watch the videos, not in how engaged they were. Although high science curious women were less likely to choose to watch the video than high science curious men, science curious men and women who chose to watch the videos were equally engaged. Furthermore, the decision to watch the video appeared to be conditional on science knowledge. Women who are high in science curiosity but modest in science knowledge were less likely to watch the videos than their male counterparts and less likely to watch than high science curious women who are also higher in science knowledge. Ongoing research seeks to understand this further, testing the extent to which stereotype threat may play a role in this phenomenon.

How do Individuals Evaluate the Trustworthiness of Vaccine Information on Social Q&A Platforms?

Speaker: Aviv J. Sharon
Technion - Israel Institute of Technology, Israel

Aviv J. Sharon, Elad Yom-Tov, Ayelet Baram-Tsabari. Despite growing concerns about vaccine hesitancy, little is known about the ways individuals seek and evaluate vaccine information online. Here, we explored the factors that predict the perceived trustworthiness of online sources. Specifically, we asked what kinds of questions are asked about vaccines, how often they are directed at experts or parents, and what features of the answers predicted answers' perceived quality and trustworthiness. We analyzed 4,910 questions and 2,583 answers retrieved from two social Q&A platforms: "Yahoo! Answers" and the Facebook group "Talking about Vaccines." Quality was assessed based on the "best answer" feature and trustworthiness was measured using external ratings of answerers' competence, integrity and benevolence. Findings indicated that on different platforms, vaccine-related questions focused on different topics; questions on one platform focused on the risks and benefits of vaccination, whereas on another, they focused on the vaccine schedule. On both platforms, most questions did not explicitly ask for professional expertise or parents' experience. On "Yahoo! Answers," both pro-vaccine and anti-vaccine answers were proportionately represented among "best answers." However, if an answer was written by a health professional, the askers and the community were twice more likely to choose it as the "best answer," irrespective of its stance towards vaccination. By contrast, an online experiment involving 694 raters of 600 distinct online answers revealed that an answer's perceived trustworthiness mainly depended on the evaluator's stance towards vaccines, and only weakly on answerer expertise. These findings reveal that while askers seldom call for expert answers explicitly, they rate expert answers more favorably. However, readers with low stakes in the assessment (paid raters) prefer answers that confirm their preexisting stances. The findings highlight the importance of open-mindedness and trust in mainstream medicine with respect to vaccines.

Scientists as Comedians: The Effects of Perceived Humor on Perceptions of Scientists and Scientific Messages

Speaker: Sara K. Yeo
University of Utah, United States

Sara K. Yeo. Humor is an important conduit for public engagement with science and is often recommended for scientists looking to conduct communication activities despite relatively little empirical evidence demonstrating its effectiveness. Here, we examine the social environment of a joke through a two-condition experimental design that manipulates the presence or absence of audience laughter. Specifically, we examine how perceived humor from viewing a video clip of a science comedian embedded in an online survey can have downstream effects on whether people view comedy as a valid source of scientific information. We found that respondents who perceived more humor in the video clip (i.e., those in the condition with audience laughter) had more positive views about comedy as a valid source of scientific information. Interestingly, this relationship was mediated by perceived expertise, not likability, of the scientist engaging in comedy.

Laypeople’s misconceptions of the tentativeness and credibility of novel scientific findings

Speaker: Danny Flemming
Leibniz-Institut für Wissensmedien, Germany

Danny Flemming, Joachim Kimmerle. Laypeople, that is, non-experts in a certain scientific domain, have better access to novel scientific findings than ever before. One important source of information are science journalistic articles on the Internet. Novel scientific findings reported in such articles are always tentative to a certain degree, that is, they can be one-sided, blurred, incomplete, or quickly outdated due to constant further development of research (Bromme & Goldman, 2014). Recognizing this tentativeness, distinguishing "correct" from "incorrect" knowledge and dealing with seemingly contradictory findings is by nature particularly difficult for laypeople. It is also affected by various text-inherent factors (e.g., the salience of the tentativeness, the presence of conflicting information) as well as intrapersonal factors (e.g., people’s attitude, epistemological beliefs, or self-efficacy) (Flemming et al., 2015). In addition, there is a seemingly paradoxical empirical finding: the higher laypeople perceive the tentativeness of research findings, the lower they rate the perceived scientific credibility of the text reporting these findings (Flemming et al., 2015). In other words, the better laypeople are informed about the limits of the reliability and validity of research findings, the less scientifically credible they consider what they are reading. Obviously, laypeople hold misconceptions about the tentativeness of scientific findings. One promising approach to resolve these misconceptions is the adaptation of refutation texts. The planned contribution summarizes empirical findings on the phenomenon and gives an outlook on possible solutions for this problem.

Trust, knowledge or online use: What shapes people’s problem awareness about climate change?

Speaker: Anne Reif
TU Braunschweig, Germany

Anne Reif, Monika Taddicken. Today, the Internet as important source of science information offers new opportunities for gaining information about and engaging with science (O’Neill and Boykoff, 2011). For climate change (CC), there are two opposing trends: (a) On social media, even influential politicians claim CC to be a ‘hoax’. On contrast, (b) young people worldwide use social media to organise and strike for a forceful climate policy. Thus, while some online users seem to lack trust, others strongly believe in the truth of scientific knowledge. In line with that, Taddicken & Reif (2016) argue with different groups of media users. Correlations between media use, knowledge, trust and attitudes may vary across these groups.

The two aims of this study are to (1) analyze how media use, knowledge about CC and trust in the climate sciences are connected to problem awareness and (2) whether and which different effects can be observed for different groups of engaged online users.

Data from a German online user representative survey (n=1,463) was used for an OLS path regression model. The results (1) indicate that trust is the strongest predictor of problem awareness. A more frequent use of online media use is connected to higher trust despite negative effects on knowledge. These effects, however, differ for diverse groups of engaged online users (2). The participating experts (highest online engagement, high procedural knowledge), for example, have the highest trust in the climate sciences despite perceiving scientific results as extremely uncertain. The more they use online media on CC, the higher is their knowledge, without effects on trust. Contrary, for the group of less active unknowing, online use is negatively linked to trust. Which specific information were used by the respondents, however, is unknown, which may be promising to be studied by future research. 

Return to:   Session parallels: draft program   |   Visual presentations: draft program