Skip to content

Can science journalists not read science papers?

Tonight I was checking my Twitter feed when I found this blog post on Nature Soapbox Science blog. Where it says that, at a Royal Institution debate about science and media

One of the questions that came up was whether reporters should read the scientific papers related to the story that they are covering.

Is that even a question?

Perhaps it’s me being naive, but I honestly thought that every professional science journalist (that is, someone who is devoted to write about science, and not just some random journalist who happens to have to write about science) had to have a science degree (and possibly some research experience). I couldn’t think someone could enter this job without being able to follow, at least generally, a scientific paper.

I am honestly scared that a serious conference, and then a Nature blog, no less, both feel this is a genuine question to be asked.

It’s true that most scientists do not read beyond their own discipline, but that’s exactly what should separate a potential science journalist from a normal scientist. A science journalist should be someone who enjoys reading papers from a lot of disciplines -that’s what the job is all about!

At least, reading and learning about multiple disciplines was one of the factors that led me to think I was suited to be no more a scientist but a potential science writer (in Italian, in case you wonder about my shoddy English): I was spending more time reading papers about entirely other stuff than work-related ones.

Yes, jargon can be a barrier between different disciplines -and surely if you give me a particle physics paper, my molecular biologist self can be confused. Still, again, a science journalist should be someone who knows a bit of every field, and if they meet some concept they do not know, they should do their best to learn it, at least roughly, before going on writing. Science writers are there to let people understand science: how can they explain to others if they do not understand it themselves?


  1. I’m pretty sure that most science journalists who write about dark matter can not provide a cogent explanation of why it may or may not be baryonic.

    I’m pretty sure that most economics writers can’t tell you where the peak of the Laffer curve has been determined to be, either.

    Accordingly, I am not at all surprised that the question came up. I agree with your opinion on what the answer should be. So what can we do about it?

  2. It’s quite obvious that a science journalist can’t be an expert on everything. However she/he should have enough expertise to be able, at least, to try and learn about the subject as much as possible. And to do that in any meaningful way means you have to have at least some scientific background, which helps even in a subject distant from the one you originally have expertise in.

    I am not a physicist for example, but I have enough background to understand -at least at a rough qualitative level- what a baryon is, and therefore to get that when we talk about baryonic dark matter we mean dark matter made of common nucleons. In turn, it becomes easy from that to understand, again at least qualitatively, why what we know about big bang nucleosynthesis puts limits on the amount of dark matter.

    But that’s even the least of problems. The worse problem of a non-expert journalist is the lack of knowledge in how science is done practically, and therefore such a journalist is helpless when scientists throw exaggerated claims or babble nonsense as sometimes unfortunately happens.

    What can we do about this? Well, I honestly do not know. At least I would like to know, when I read a science article on a newspaper, what are the credentials of the journalist. This could be a little start.

  3. It’s true to say that the number of baryons depends on nucleosynthesis. But it’s technically more correct to say that quantity depends on the density of the universe during nucleosynthesis, and that density in turn is highly sensitive to the rate of the metric expansion of space between the inflationary epoch and nucleosynthesis — a rate about which we have absolutely no information. All we know about it is that the universe expanded very quickly during the very brief inflationary epoch, and then slowed down for several minutes afterwards until nucleosynthesis began.

    We don’t know how much the universe slowed down in those minutes, and even the fact that we don’t know the resulting rate and density has apparently been lost to the sands of time. Most astrophysicists don’t know that the density is indeterminate, but the linear extrapolation of the rate of expansion to the present has been the only value in use for so long that the idea of baryonic dark matter has become a heresy. Even though we know the expansion rate slowed down after inflation and is speeding up now.

    But, there is hundreds of millions of dollars of grant money paid annually for proposals to build nonbaryonic dark matter detectors, even though only one lab has claimed detection and all attempts to replicate their work has failed. So maybe the fact that we don’t know the rate of expansion prior to nucleosynthesis can remain forgotten for the sake of big science.

    Similarly most economists have no idea that the peak of the Laffer curve has been measured at about 84%, because that is a very radical value politically. Most economists have seen pictures of the Laffer curve with peaks around 50%, and conservative politicians constantly repeat the assertion that the peak is below the current tax rate, no matter how low it goes. In the mean time inequality skyrockets, with 1.6 million homeless children in the US up from 1.2 million just three years prior.

    Where did this mistake come from? A math error. A mistake made by Arthur Okun in 1975, 37 years ago:

    I would rather have the opportunity to read the paper being reported on, or at least its abstract, than know about the credentials of the journalist. I have lost all faith in the competence of people with credentials.

  4. Your comment is intriguing. First of all, your last point about having the actual paper in hands is very important. I personally strive to always link the full paper or, when it is behind a paywall, the abstract. And this also brings the fundamental issue of open access -that is, science results should be freely and openly available to the public at large in their unadulterated form.

    About the rest, well, I’m not that informed to comment in detail, but I am always wary of claims like “most astrophysicists don’t know”. It is statistically implausible that a whole scientific community ignores such a vital detail and you, who happen to be outside such a community, knows it. It may be so, but 99% of the times such claims arise from misinformation by the outsider.

    Also your second link is absolutely interesting and enlightening (and thanks for it!) but it doesn’t talk of Laffer curve nor trivial math errors as you seem to imply -it simply says that further research has revealed new insights about the relationship between inequality and growth. That is the normal scientific process, improving and fixing previous knowledge. That the political world at large ignores that is a very different problem.

  5. Your comment is a nice illustration of the abyss between scientists and journalists. You begin by this question, should they read scientific papers, and in the same paragraph, you jump on the fact that science journalists should have a degree in science.

    Yes, the question about the papers is scary. Yes, science journalists should read the papers (most of them do). But on the other hand, the possibility that it would be mandatory for science journalists to have a science degree, is scary too: sometimes, to be critical about something and to explain it clearly, you have to take some distance.

    So, this post of yours show how important it would be for scientists to better understand how journalism work. Science journalists do make this effort —because this is part of their work. If scientists had the same understanding about how the medias are working, a lot of misunderstanding could be avoided.

  6. To me it’s a very simple matter: if you explain something, you should know (at least broadly) what are you talking about.

    A science journalist is not so different from a teacher: both are people explaining non trivial concepts to non-experts. We ask teachers to understand their subject -why don’t we ask the same to journalists?

    That media currently work ignoring this simple issue doesn’t mean it is right for media to work this way. One thing is “to take some distance”, another is to be unable to understand scientific concepts but feeling entitled to explain them to the public nonetheless.

  7. Jonathan Jonathan

    I don’t think journalists should always read papers; in many instances, interviewing a researcher is just as good or better. Often the impact of a paper can only be understood within a broader research context. Understanding that context takes more time than a journalist typically has. An interview can offer the same quality information, from the same source, in less time.

    Also, I want to mention an argument in favour of journalists without scientific education: someone who doesn’t automatically understand the more basic principles of a field will aks questions about those principles, and thus enquire about things that a lay audience may want to know.

  8. An interview can offer the same quality information, from the same source, in less time.

    It doesn’t. For example scientists, when interviewed, tend to overestimate and oversell the impact of their own research. It happens all the time, and I’ve seen it happening myself in more than one occasion. A very well known British scientist, which I worked closely with, once in an interview claimed that his research was going to find a cure for Alzheimer and Parkinson in the next 3 years. This was and is utterly, completely, ridicolously false: as we researchers clearly understood (and as anybody reading his papers could immediately understand). Yet that’s what he said. You can’t lie too much about the impact of your stuff in a peer-reviewed paper, but you can lie as much as you want to a non expert journalist.

    someone who doesn’t automatically understand the more basic principles of a field will aks questions about those principles

    I agree you need to think like a non-expert to understand what people need to know. However an expert (broadly construed) can at least try to think like a non-expert; viceversa is not possible.

  9. Jonathan Jonathan

    “Scientists, when interviewed, tend to overestimate and oversell the impact of their own research.”

    That’s not the journalist’s fault. If what you say is true as a general rule (which I doubt, especially given the sample size you mention), then it still is a terrible reason to let the papers take precedence in science journalism.

    My point is certainly not that they never should – far from it, as far as possible, journalists should check the actual research articles. But assuming that interviews are inadequate because scientists will overestimate the impact of their results is denial of what is part of the social impact of science: the scientist’s motivations and interpretations of her own work. I say publish the overestimations, because they are relevant.

  10. Your comment is a perfect example of why science journalists should have science experience.

    If what you say is true as a general rule (which I doubt, especially given the sample size you mention),

    It is a, say, 50% rule. Yep, I mentioned only one example, but at least half of group leaders in the world will tend to exaggerate their claims in interviews and press releases.

    That’s because of a mix of factors. They honestly believe that perhaps (scientists tend to be overconfident often -otherwise surviving the job is very difficult). But most importantly getting media attention and emphasizing the importance of your work is going to get you funding, directly or indirectly. And scientists compete for funding very hard.

    then it still is a terrible reason to let the papers take precedence in science journalism.

    You got the logic wrong. It is not a direct reason to let papers take precedence. It is a reason to be very wary of using interviews as a proxy for the paper’s contents.

    Don’t get me wrong: interviews to scientists are awesome and there should be more. They can give tremendous insights and angles. But they can’t be the only information source for a science story about a breakthrough.

    what is part of the social impact of science: the scientist’s motivations and interpretations of her own work. I say publish the overestimations, because they are relevant.

    Sure, but you should be aware they’re overestimations. And how can you, without knowledge of the real data, the context etc.?

    Again, for me it’s a simple and general matter: if you talk about it, you should understand it.

  11. I’m a journalist.

    I’ve trained 10 years to process information logically, talk to people in ways that get them to tell me what I need in ways I can quote. I do not have a PhD in molecular biology, and as such, the papers that represent cutting edge developments in the field are beyond my scientific understanding.

    I skim the papers I write about and ask a lot of questions of the authors. This is the accepted practice, partly because our job is to make the science understandable to a general audience, so it’s OK for us to know more than the readers but less than the scientists.

    I do think science journalists should get some (which is to say more) training in the fields they cover. But we can’t have 2-3 PhDs, especially when we make, on average, less than administrative assistants.

    And because journalists do write for a general audience, those who specialize in the area being covered will always, by definition, find the coverage lacking. But that’s OK — you can go and read the paper yourself, having learned about it through a science journalist.

  12. I am not saying you should have a Ph.D. in every field you write about (I don’t, either). But you *should* look at the paper and do your best to understand it. And if you don’t understand it *at all*, then perhaps you shouldn’t write about it, or at least you should honestly admit it in articles.

    Asking questions to the scientist is not the same, not at all. They’re not necessarily great communicators to the public and their explanation may be misleading. Plus, they will sometimes try to oversell their work. Which is incidentally good for the story (everyone wants the story to look important), but less good for accuracy.

  13. […] is, the original technical scientific papers describing a result. It might sound surprising but many science journalists actually think that they can write about the science without even having a …To their peril, however. If the journalists who today prattle about “Earth 2.0″ actually […]

Leave a Reply

Your email address will not be published. Required fields are marked *