science journalists may be expressing greater uncertainty in the coverage of research studies than the researchers themselves, according to a new study from
the university of michigan
.
researchers compiled a collection of 128,942 u.s.-based news articles and the abstract of the research paper they reported on. these articles were from 273 different news outlets covering 57,807 different scientific articles. by comparing the language and context in each abstract/news article pair, the researchers indicated that, in general, news used language and sentence structures that communicate less certainty than the research articles.
“the findings presented in the science news are actually lower than the certainty of the same scientific findings presented in the paper extracts,” jiaxin pei, an author on the paper,
told poynter
.
examples of using uncertain language when reporting on a research study include using phrases like “we
suspect
a has effects on b,” versus “we
conclude
that a has effects on b.” other words, like possibly, maybe, and approximately, also expressed a degree of uncertainty.
the researchers noted that only looking at the abstracts of these studies is a limiting factor. more research into the actual study is needed to get a larger picture of the differing ways science is being communicated.
society is biased toward results
however, not all researchers are convinced that reporting on news stories is entirely non-sensationalized. a case study published in the
journal of science communication
looked at the reporting from australian news agencies on a single research article as an example.