Daily Reads: No Comment

Today’s Daily Read isn’t the most scintillating article I’ve ever read, but the information it contains is revealing and important, so I believe it is worth the quick read. Cathleen O’Grady of Arstechnica reports on two studies that looked at the influence of online comments on the effectiveness of public service announcements (PSAs). The study created mock PSAs on the safety and efficacy of vaccines – one pro-vaccine and one anti-vaccine – and paired them with anonymous comments that supported or refuted the message of the PSAs. The researchers made sure to attribute each PSA to health organizations that were perceived as credible by the study participants. In the first study, participants were more likely to have their opinions swayed by the PSA if they thought the source was credible; however, if they found the commenters to be credible they were less likely to be swayed by the PSA. In the second study, the participants were provided with general information about the commenters – e.g. one commenter was identified within the comment as a college student, another as a health care lobbyist, and a third as a medical doctor specializing in disease and vaccines. In this study, the participants were more swayed by the commenters they perceived as credible – in this case, the self-identified doctor – than they were by the PSA. This occurred regardless of whether the doctor agreed or disagreed with the PSA. As the article notes, this research is not definitive, but it points out an important weakness in how people perceive information they receive from anonymous comments – which is, how can we know that a self-identified commenter truly is the expert he or she claims to be? I think this is hugely important, and very troubling. It’s one thing to trust the credibility of an anonymous restaurant review on Yelp; it’s quite another to trust anonymous opinions on critical health and social issues.

Don’t read the comments—they can make you mistrust real experts

Leave A Comment

Your email address will not be published.