An opinion piece in the journal Science (paywall) suggests that readers are influenced by online comments on a story as much as they are by the story itself, according to an article by Mark Johnson in the Wisconsin Journal Sentinel.
Researchers at the University of Wisconsin said they did a study in which they prepared a balanced news report about nanotechnology and showed it to two groups of readers. One group saw civil comments; the other saw uncivil comments and name-calling. "Disturbingly, readers' interpretations of potential risks associated with the technology described in the news article differed significantly depending only on the tone of the manipulated reader comments posted with the story," wrote the study's authors, Dominique Brossard and Dietram A. Scheufele, according to Johnson's story.
Johnson isn't entirely clear about how readers' views of risks changed with the uncivil comments. In the Science paper, Brossard and Scheufele write, the uncivil comments "polarized the views among proponents and opponents of the technology with respect to its potential risks." That's not clear, either. Did those who read the uncivil comments become more, or less, afraid of the risks of nanotechnology? A report of the study will appear in a forthcoming issue of The Journal of Computer-mediated Communication.
This strikes me as a potentially important development. With the exception of Johnson's piece and a brief story at Poynter.org by Andrew Beaujon, I found no coverage of this report. I don't know why it didn't receive more coverage, and I don't know why the coverage--and the article in Science--were not clearer about precisely what the researchers found.
The University of Wisconsin news release was clearer: "The less civil the accompanying comments, the more risk readers attributed to the research described in the news story," it said.
If that's true, bloggers and others who cover science online might want to change the way they treat comments. But the idea that managing comments can change readers' perception of a story opens the door to manipulation of comments in a way that we might find objectionable, even dangerous.
I'm looking forward to a broader discussion of this finding, perhaps when Brossard and Scheufele's study appears in the communication journal.
Meanwhile, the science press, which should be watching this closely, mostly ignored it. I don't understand why.