June 12, 2024

How to Evaluate Trustworthiness in Science

Illustration of online doctor giving advice. Shutterstock / Katy Flaty

These days, anyone can write or talk about science and health topics, no matter how much or how little聽about the topic they actually know. Some might have a deep understanding of the topic but not be great at talking about it. Others could have little knowledge about a topic but sound very confident and persuasive.

Even well-trained scientists can find it difficult to sort out contradictory information and evaluate trustworthiness in science from a field that鈥檚 not their specialty. So, when it comes to health and science topics, how can you evaluate the source of information? Here鈥檚 a list of questions we鈥檝e put together that you can ask yourself:

  • Is the person an expert in the area they are discussing? Researchers in one field aren鈥檛 necessarily experts in related areas. What are their credentials?
  • Is the person using emotional language that could influence your opinion or judgment?
  • Whose interest is the person serving? Is there any reason they might have a bias? For example, are they trying to sell something like a book, product, or treatment? What are their associations or affiliations? Do they have a political motivation?
  • Does the person show that they understand the nuances of research or are they speaking in absolutes? Research findings are seldom so clear there aren鈥檛 any caveats.
  • Does the person show an understanding of the broad context, or are they cherry-picking鈥攃hoosing only studies that support their point of view or highlighting only certain information? Are they considering other perspectives and interpretations, or are they simply dismissing them?
  • Is the person challenging a known scientific consensus? What do other experts in this field say about the subject?聽In science, challenging the current consensus can change the way people look at problems in important ways. But in trying to push a particular point of view, challengers could also be ignoring large amounts of data that might prove them wrong.
  • Are numbers being presented accurately and appropriately or are they being manipulated to support a particular point of view?
  • Are health risks put in perspective? Is it clear who a risk affects, where, when, and how certain the risk is? Does the person imply that everyone should be concerned about the issue when only certain people are at risk?
  • Is the person talking about a body of research or only one specific study? If it鈥檚 only one study, consider:
    • Was the study done in humans? If so, what are the attributes of the people who were studied (such as age, sex, ethnicity) and who would the results apply to?
    • What kind of study is it? Was it designed to answer the research question the conclusions describe?
    • Was the study size appropriate to answer the question?
    • How old is the study? Is this a new study that still needs to be replicated? Or have others replicated the findings?
  • Has the work been reviewed by other scientists (published in a rigorous peer-reviewed scientific journal)? Or is the report based on a preprint or a talk or poster at a meeting?
  • Did the researchers engage the community or populations impacted by the issue to make sure they鈥檙e considering factors they may be unaware of?

Your answers to these questions can go a long way in helping evaluate the quality of information for both you and your readers.

We welcome your comments, advice, and additions. We hope to hear from you, whether you鈥檙e new or an experienced communications professional. Please share your thoughts by sending an email to聽sciencehealthandpublictrust@mail.nih.gov.

Links/Resources

  • 听(狈础厂)
  • (NAS)