New study highlights complexity of sensationalism in science reporting

As I’ve touched on before, sensationalism in science journalism is a really tricky issue and some Australian research about sensationalism in health reporting released today further complicates the issue, revealing that reporting of the causes of cancer is often very misleading.

On the one hand sensationalism is an important part of communicating scientific achievements to an intelligent but non-expert community. It is, in part, a matter of focusing on the relevance some particular discovery has to people’s lives.

For example, an article I recently wrote about the discovery of monopoles mentioned near the very top that it may have drastic implications for personal computing. This was something hardly mentioned by the researchers but it seemed like something one ought to mention to a lay audience since it is true and it will pique their interest in the important discovery.

On the other hand, even this innocuous kind of sensationalism does some disservice to the lofty aims of science communication. Focusing on issues that were almost irrelevant to the researchers misrepresents the way science progresses, the motivations behind research and downplays the implications it has for our theoretical understanding of the world.

In other cases, the requirement for sensationalism in science communication and reporting leads to quite harmful outcomes.

Professor Bernard Stewart from the University of NSW Faculty of Medicine and South East Sydney Illawarra Health studied 100 media reports of the causes and correlates of cancer and found that many of them encouraged fear of products based on little or no evidence of their link to cancer.

“Many of the reports headlined unrecognised carcinogens or suggested carcinogenic risk, yet this was not the focus of the original research findings,” Professor Stewart says. “Consumers were given the impression they needed to be overly vigilant, though cancer causation had not occurred and often seemed very unlikely.”

Stewart doesn’t propose a simple-minded campaign against sensationalism but instead recognises the tremendous complexity of the problem.

Pointing to the various interests at play in getting a story in the media, Stewart compares the process to a game of Chinese-whispers. “From the media release written by the institutional press officer, to the journalist trying to simplify often complex research, to the sub-editor coming up with an attention-grabbing headline, the message is gradually altered to centre upon the likelihood of cancer.”

In an article I published last week, I made the same analogy and pointed to another, perhaps more problematic, part of the game: Once the story is in one media outlet, it is often picked up by other outlets without any original research and further sensationalised and often infused with further inaccuracies.

Also, it should be noted that this kind of process doesn’t fully explain the worst kinds of sensationalism that we see around drug reporting. When it comes to reporting on the relative risks of alcohol and other drugs, the misrepresentation of the facts is so clear and complete that some other explanation is called for.

According to the press-release, “Professor Stewart said journals should be more responsible in publicising investigator speculation, journalists needed to ensure novel findings were placed in context, and sub-editors needed to take care not to inadvertently sensationalise preliminary observations.”

This is all good advice but how one might institute these recommendations is very unclear.

Professor Stewart will present his research at 4.40pm today (Nov 17) at the Clinical Oncological Society of Australia Annual Scientific Meeting. Meeting room 8, Gold Coast Convention Centre.