The good news from Adrian Barnett and colleagues is that scientific research writing is getting clearer.
The bad news is “many scientists” still use “The Official Style,” “that falls back on dense and difficult prose.”
To identify the impact of writing up research in language that is big on jargon, passive language and a “detached voice,” the authors searched 20 million abstracts over 70 years for nine key writing components, “known to aid reader engagement.”
They found improvements:
- starting in the ‘50s, less hedging and more active narration. “This indicates that researchers are putting some effort into writing in a more reader-friendly manner”
- shorter sentences (average 29w in the ‘60s and 15 now).
But also worse things; notably noun-chunks (three or more in a row), too many numbers and “sensationalistic words.” These come in nine varieties, including, words communicating novelty, importance and quality. In astounding news, they suggest, “sensationalism can change the way readers interpret a paper and give them misleading impressions about the results.”
Overall, however, there is optimism, which matters. “The benefits of the internet age for science are limited if our work remains inaccessible behind a linguistic wall. It is valuable for researchers to pay attention to the ways in which papers are written. In doing so, they can ensure their intended message rings clear for readers,” they write.
