While the flashy Clickbait headlines that promote the power of chocolate to cure everything from acne to cancer are certainly eye-catching, these articles may not be common in science communication.
Large-scale study at the University of Michigan Uncertainty In science communication, journalists show that they tend to soften scientific claims rather than exaggerate them.
A new study by UM School of Information scholars Jiaxin Pei and David Jurgens delves into how scientific uncertainty is communicated in news articles and tests whether scientific claims are exaggerated. They also wanted to see how the scientific claims of news differ between respected and peer-reviewed journals and less rigorous publications.
“When journalists talk about the possibility of exaggerating their claims, it always feels like these extreme cases,” said Jurgens, an assistant professor of information. “I wanted to see if there was a difference between what scientists said and what journalists said about the same paper.”
Overall, Pei and Jurgens, Science communication..
“Our findings suggest that journalists are actually quite careful when reporting science,” Pey said, with some communicators rather than journalists claiming science. He added that it reduces certainty.
“Journalists are doing a lot of work,” said a journalist who recognizes the skills needed to translate scientific results into the general public. “It’s great to see journalists actually trying to contextualize and soften scientific conclusions in a wider space.”
For their work, researchers focused on certainty that could be expressed in subtle ways.
“There are many words to show how confident you are,” Jurgens said. “It’s a spectrum.”
For example, adding words such as “suggestion,” “approximate,” and “possible” tends to increase uncertainty, but using accurate numbers for measurements increases certainty.
Pei and Jurgens obtained news data from Altmetrics, a company that tracks scientific paper references in news articles. They collected about 129,000 news articles that referred to specific scientific articles for analysis.
In each of the news articles and scientific papers, we analyzed sentences containing found words such as “find” and “conclusion” to see how journalists and scientists made their paper claims. A group of human annotators examined scientific papers and news articles, focusing on the certainty levels of more than 1,500 scientific discoveries.
“We took the claims in the summary and tried to match them with the claims found in the news,” Jurgens said. “So we said,” Okay. Two scientists and journalists are trying to explain the same thing, but they’re aimed at two different audiences. What’s in terms of certainty? can you see?”
The researchers then built a computer model to see if they could reproduce the certainty levels pointed out by human readers. Their model was highly correlated with human assessment of how certain the claim was.
“The performance of the model is sufficient for large-scale analysis, but not perfect,” said Pei, a PhD student at UMSI and the lead author of the dissertation, mainly because of subjectivity. He explained that there is a gap between judgment and machine prediction. ..
“When identifying textual uncertainty, people’s perceptions can be diverse and it is very difficult to compare model predictions with human judgment. Humans sometimes disagree with many. You may.”
Pay says that the translation of a study can be more ambiguous when it comes to the quality of the journal, or what researchers call the impact factor of the journal. Some science newswriters report a similar level of certainty in the news, wherever the original research is published.
“Given that the impact factor of journals is an important indicator of research quality, this can be a problem,” he said. “The journalist Nature Also Chemistry Also, in some unknown journals with the same degree of certainty, it may not be clear to the audience which findings are more reliable. “
Overall, researchers see this research as an important step in better understanding the uncertainty of scientific news. They have created software packages for scientists and journalists to calculate the uncertainty of research and reporting.
Journalists can benefit from checking the certainty of their work, but journalists say this tool may also help readers.
“It’s easy to be dissatisfied with uncertainty,” he said. “I think providing such a tool has some calming effect. This work is not a silver bullet, but I think it helps the reader’s overall understanding.”
This work was featured in the minutes of the 2021 conference on empirical methods of natural language processing.
This survey is available in PDF format. aclanthology.org/2021.emnlp-main.784.pdf
University of Michigan
Quote: According to a survey, journalists obtained scientific claims from https://phys.org/news/2022-02-journalists-tend-tempernot-exaggeratescientific.html on February 21, 2022 (February 21, 2022). It tends to soften the day).
This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. Content is provided for informational purposes only.
Studies show that journalists tend to soften scientific claims rather than exaggerate them.
Source link Studies show that journalists tend to soften scientific claims rather than exaggerate them.