A common complaint I hear from researchers is that journalists make a lot of mistakes when they report on academic studies. They often describe study findings incorrectly, for example, or claim that a new paper proves something when it doesn’t.
I’ve written dozens of tip sheets in recent years to help journalists fine-tune their skills in choosing, vetting, understanding and explaining research as part of their reporting process. This tip sheet, however, is for researchers, who also play a role in helping journalists get it right.
Our main goal at The Journalist’s Resource is bridging the gap between newsrooms and academia to ensure news coverage of public policy issues is grounded in high-quality evidence — peer-reviewed research in particular. Everyone benefits when journalists report accurately on research findings, especially the everyday folks who make decisions about their health and safety and their children’s futures based on that information.
When I speak to groups of researchers about the best ways to build relationships with journalists, I often share these 10 tips with them. They represent some of the many ways researchers can help the news media avoid errors, starting with the journalists who interview them about their own work.
1. Use plain language.
Many journalists haven’t studied research methods or statistics and often don’t fully understand the technical terms researchers use to communicate with one another about their work. That’s why it’s important to use plain language when discussing the details of a research paper.
For example, instead of saying you found a positive association between air pollution and dementia in older adults, say you found that older adults exposed to higher levels of air pollution are more likely to develop dementia.
Another example: Instead of saying there’s heterogeneity in the results of three experiments, you could say the trials produced different results.
2. Make sure any press releases written about your research are accurate.
When journalists cover research, they sometimes look to press releases for guidance in describing research findings and to double-check key names, facts and figures. Unfortunately, the press releases that higher education institutions and research organizations issue to promote their researchers’ work sometimes contain errors.
When possible, review the final version of a press release about your research before it’s shared with news outlets. If you spot problems after it has been distributed, ask for a correction and note the error in the initial press release when speaking with journalists.
3. Offer examples of the right and wrong ways to explain relationships among the key variables studied.
Not all journalists know what causal language is, or when they should and shouldn’t use it to describe the relationship between variables in a research study. When speaking with a journalist about a study’s findings, point out whether there’s evidence of a causal relationship between or among key variables. If there isn’t, offer the journalist examples of correct and incorrect ways to explain this relationship.
An example: Let’s say a study finds that crime rates increased in 10 cities in 2020 immediately after local police departments implemented a new crime-fighting program. Let’s also say researchers found no evidence that this new program caused crime rates to rise.
A researcher discussing these findings with a journalist could help them avoid errors by explicitly pointing out the right and wrong ways to report on them. In this case, tell the reporter it would be inaccurate to say this study finds that this program causes crime rates to rise, might cause crime rates to rise or leads to higher crime. It also would be inaccurate to say that introducing this program contributed to higher crime rates in those 10 cities in 2020.
Then share examples of correct ways to describe what the study’s authors learned: They discovered a relationship, correlation or link between this new program and increased crime rates in these specific cities in this one year. However, researchers found no evidence that the program caused, led to or contributed to the increase.
4. Note the generalizability of findings.
Many news stories and news headlines overgeneralize study findings. Researchers can help journalists get it right by noting how generalizable findings are.
Example: Let’s say an academic paper concludes that 25% of student athletes at public universities in one state reported using marijuana during the past year. A researcher doing an interview about this study could help ensure the journalist reports on it accurately by stressing that the findings apply only to student athletes at these specific universities. It would be helpful to also point out that it would be incorrect to insinuate these findings apply or might apply to other types of students or to student athletes at other higher education institutions.
5. If a study’s results are expressed as standard deviations, be prepared to help journalists explain those findings using common measurements the public will understand.
A lot of journalists will need assistance describing results reported as standard deviations. Mainstream news outlets will generally avoid the term in news stories because it’s unfamiliar to the general public. Also, even when it’s explained using plain language, the concept can be difficult for even the most educated audience members to grasp.
One way to help audiences comprehend results expressed as a standard deviation is by describing the results using more common units of measurement such as points, percentiles, dollars and days. Santiago Pinto, a senior economist and policy advisor at the Federal Reserve Bank of Richmond, does a good job explaining standard deviation changes in student test scores in his August 2023 report, “The Pandemic’s Effects on Children’s Education.”
The report looks at how U.S. eighth-grade students performed on the National Assessment of Educational Progress between 2019 and 2022. Pinto writes:
“What does a decline of 8 points in the NAEP math test mean? In math, one SD [standard deviation] of individual student scores is about 40 points and is roughly equivalent, as a rule of thumb, to three years of schooling. The national average loss of 8 points is equivalent to 0.2 SD, which implies 0.6 years of schooling lost.”
6. Ask journalists to briefly summarize the key takeaways of your interview with them.
Doing this at the end of an interview is a good way to gauge how well a journalist understands the research. And it allows you to correct any errors and misunderstandings. Keep in mind, though, that journalists working on a tight deadline will have limited time to go over the key points of your conversation.
7. Offer to answer follow-up questions and review word choices.
At the end of an interview, invite journalists to contact you if they have additional questions, including questions about whether they’ve explained something correctly in their story. Journalists generally won’t share copies of their work before it runs — news outlets tend to discourage or prohibit it. But they might share a few sentences or paragraphs when they ask for help making sure there aren’t mistakes in those parts of the story.
If you’re worried about being quoted correctly, ask that they read back any direct quotes they plan to use. Some journalists won’t mind doing this, provided they’re not on an immediate deadline.
When you offer to answer follow-up questions, point out the best ways and times to reach you.
8. Provide examples of accurate news coverage and summaries.
You can also help journalists check their work by sharing news stories, reports and summaries that correctly characterize the research the journalist is reporting on.
9. Share The Journalist’s Resource’s tip sheets.
You’ll find a link to our “Know Your Research” section on the right side of our homepage. We’ve created tip sheets, explainers and other resources to help journalists build research literacy and numeracy. Our tip sheets cover topics such as statistical significance, standard deviation, the purpose of peer review and how to interpret data from polls and surveys.
All our written materials are free. We publish them under a Creative Commons license so anyone anywhere can share and republish them as much as they like.
10. After the news story runs, give feedback.
Good journalists want to know if they got something wrong. They correct their mistakes and try to learn from them.
Too often, when researchers spot errors in news stories, they do not alert the journalist. If no one raises an issue with a news story, the journalist who reported it — and all the other journalists who will use it as a reference in the future — will assume it’s accurate.
Likewise, tell journalists when they’ve done a good job covering research. One way you can let them know they got it right: Share their story on social media. Also, reach out with new research you think they’d want to know about.
Expert Commentary