My bad – on accountability in science journalism

By Ed Yong | March 6, 2012 9:05 am

There are many reasons why errors creep into science journalism, beyond what the journalist happens to write. Scientists can publish fraudulent papers, draw false conclusions from their experiments, or give misleading quotes. Press officers can paint results in a skewed light, omit or strip out context, and make up quotes entirely. Editors can assign reporters to tenuous stories. Sub-editors can introduce errors.

You already know this. I’m explicitly pointing it out because I have seen several instances where journalists have attempted to defend shoddy reporting – either by themselves or their colleagues – by citing this laundry list of reasons as mitigating factors.

The latest prominent example comes from no less than Paul Dacre himself – editor-in-chief of the Daily Mail. Ananyo Bhattacharya from Nature has a good piece in the Guardian where he singles out Dacre’s attempt to defend a piece of indefensible “reporting” in the Mail. The story, by the unfortunately named “Daily Mail Reporter”, claimed that switching on a light for a night-time toilet visit could “cause changes that might lead to cancer”. Testifying at the Leveson enquiry, Dacre argued that the story was legitimate because it was reworked from wire copy that included a quote from one of the scientists involved.

Bhattacharya writes, “[Dacre] is wrong to imply that it constitutes acceptable journalistic practice. It does not.” Big tick.

There are many reasons why errors creep into science journalism, but as far as the accountability of an individual journalist is concerned, they are all irrelevant. They are not mitigating factors. If we write something, and we put out names to it, the buck stops with us. If there is a mistake, it is our fault.

If the paper was rubbish, if the peer reviewers missed something, if the scientist lied, if the press release is distorted, it’s still our fault for producing something that is inaccurate or that fails to root out these problems.

We are meant to be the final filter that parses information for the reader – the final bastion of accuracy. We’re meant to do the necessary background research, fact-checking and reporting that exposes dodgy information. We’re meant to be immersed enough in our beats to have finely tuned bulls**t filters.

If this sounds like an unbearable line to draw, well, it is. The job’s not meant to be easy. But I am not saying that journalists should never make mistakes – that would be deranged. We have neither the time nor the level of expertise necessary to guarantee an error-free oeuvre. We will all screw up. What matters is that we exercise due diligence in getting our facts right, own our mistakes when they happen, and fix them if we can.

When we screw up – when, not if – it doesn’t matter who our accomplices in error were.

It does not matter that we didn’t introduce a mistake, simply that we let it slide.

It does not matter where errors crept into the chain. It matters that they did not end with us.

Next week, Ananyo and I are taking part in a discussion at the Royal Institution about how to improve standards in science journalism. There are loads of obvious things that one could name, and the Guardian is canvassing suggestions. To me, developing a proper sense of accountability is an important one.

As they say, the first step is admitting that you have a problem.


Comments (21)

  1. Totally correct. And, of course, this applies to all journalism, not just science journalism.

    However, most of the Daily Fail isn’t journalism.

  2. Ship

    Amen, to everything you wrote here. But, worth noting, this doesn’t absolve PIOs, et al., of guilt if they botched things. Sometimes there’s enough guilt for everyone.

    Of course, I was raised Catholic, so I would think that.

  3. Nice piece, Ed.

    I agree that accountability is pivotal. This is something we’ll be addressing in tomorrow’s Guardian piece and at the RI debate: that, as scientists, we need to assume full responsibility for our own press releases and other PR material.

    It seems to me that if both sides adopted this ethos, and at multiple points in the chain between scientist and media consumer, we might be able to eradicate considerable chunks of inaccurate science news.

  4. @Richard – Indeed, this does apply to all journalism.

    @Ship and @Chris – You’re both absolutely right that other links in the chain need to assume responsibility for their part too. This piece was motivated by the regular buck-passing I see, as if there’s only ever one person to blame for any piece. Often it’s a chain of fail.

  5. Agreed for all journalism….and about the difficulty of the job and the inevitability of errors. Journalists must commit to owning their mistakes–and then learn from them and apply the new learning to the next story and the next.

  6. Aurora

    I agree with you all, but have a bit of a quibble with Chris’ assertion that scientists have to assume full responsibility for press releases – that would risk letting press officers (like myself) off the hook. I think ideally scientists have to work *with* their press officers to ensure the release (or other PR effort) is accurate and doesn’t make outlandish or over-hyped claims, *and* still does its job of raising interest in the research.

    The other important point I think Ed touched on was fixing our mistakes when we can – we must not only fix them to appease critics and be able to point and say “see, I corrected it afterwards”, but also make an effort to alert people who have already been exposed to the mistake. And this is where I see institutional politics (both in media and in science institutions) sometimes making things worse…

  7. This is in general an excellent reminder, but as I commented on Twitter, there is a point of trust that cannot be the journalist fault. That is, a journalist can’t be expected to discover high-profile and well made scientific fraud. We can detect, hopefully, if a scientist is sloppy, if a research is not to be completely trusted, if conclusions aim too high, etc. But we have to take, ultimately, the word of the researchers for it -we weren’t in the lab doing the experiments.

    That is, we can’t ask journalists to be responsible if they report the next Jan Hendrik Schön.

    Also, it is possible to report wrong research which is perfectly honest. Even statistically, if papers use a 0.05 confidence level (like is often in biology) to draw conclusions, this means 1 paper out of 20 is wrong only because of statistics. This is literally nobody’s fault.

    When it happens, surely it should be promptly corrected and people should be fully alerted, like Aurora says. And this is a journalist responsibility. But a journalist can’t be held accountable because a paper is wrong.

  8. @Aurora – Yes I agree completely that scientists need to work more closely with press officers (and by that I mean more closely than many currently do). However I also believe that the ultimate accountability for poor-quality press releases lies with the scientist, not with the press office.

    The reason I take this view is that the scientist is best placed to know if a particular claim in a press release is an exaggeration, or if a key limitation should be highlighted. It isn’t fair to expect press officers to be cognisant of these details, so if a scientist remains silent and allows an inaccurate press release to be issued, it isn’t fair or logical to blame press officers.

    As Ed says, it’s ultimately a chain of shared responsibility, but if scientists took a greater role in monitoring their own PR, and as you say, worked more closely with press officers to maximise impact and accuracy, we could make strides in improving the quality of the information that we feed into the hungry media machine.

    (btw, we’re also planning to do some research on this exact issue – see our website…and don’t hesitate to drop us a line if you have any suggestions)

  9. Having disagreed with @devicerandom on Twitter, I actually agree with much of his comment. I think the last point is important though – if we do cover fraudulent research, it is our responsibility to correct and alerting.

  10. Yep, I fully agree that there should be a much higher profile to correcting and alerting.

  11. My hat off to you kind sir! Accountability goes a long way towards integrity.

    I don’t mean to rub it in, and I hope you are candid enough to appreciate me bringing it up again as an example of where our biases get in the way of better judgement, but I feel that the circumstances were peculiar enough to warrant mention anyway: I was totally baffled by how uncritically you and a lot of other popular science writers handled the infamous viral statement that popped up to exonerate the engineering of Fukushima.

    There I really would have hoped for many writers to step up and check in hindsight why they uncritically helped a story to go public that was of dubious origin. It is in those helter skelter moments of public shock and PR-machines kicking into overdrive that your expertise in fact checking and knowing your beat is most sorely needed, and needed fast in the age of social media.

    I actually do have some understanding of the sociology at work and the biases (like sharing a positivist world view for example) that some messages feed into, I respect the feelings involved that led to the communication fiasco that was Fukushima and even wrote about it:

    Yet I really would hope that perhaps you would address the issue of creating or buying into narratives at this conference of yours and how one may mitigate this. Even more difficult, how to maintain composure when the shit really hits the fan in the twitterverse, perhaps?


  12. Aurora

    @Chris: I wholeheartedly agree that scientists should take serious interest and work closely with press officers (while still recognising that the latter have greater expertise in *how* to communicate – but that’s a discussion for another time).

    On top of that, though, I’d say it’s press officers’ responsibility to make sure scientists *do* check for errors, exaggerations, or missing limitations. There’s sometimes a temptation to think “the scientists didn’t comment, so it must be OK (and I can move on to my next task)”, or to get exasperated by a re-written text and view every requested change as an attempt to insert more jargon. As press officers, it’s our job to weed through responses and work out – with the scientists – if and where there are inaccuracies, misrepresentations, exaggerations, etc. Or ideally, to gauge and anticipate some of those issues before even drafting the text – by talking to the scientists and understanding not just what they say but how they say it.

  13. Bob Ward

    Ed, this particular case from the Daily Mail all swings on the quote it included from the lead researcher on the academic paper, which mentions the toilet light switch. This quote , the Mail claims, was contained within the wire copy that the Mail plagiarised. To establish who is originally responsible for the misleading nature of the article depends on whether the researcher actually said what he is quoted as saying (nobody appears to have checked this). But assuming that the quote is accurate, what would you have expected the reporter to do? Assume that the researcher is lying and totally ignore what he has said? Was there any reason to disbelieve what the lead author on the paper said? It seems to me that the reporter might simply have been reporting in good faith what the researcher said to him/her. What exactly would you have done differently?

    I think the fundamental flaw with these discussions is often a misdiagnosis of the problem. There should be more consideration of how researchers and their press officers are the root of many examples of bad reporting.

  14. @Bob – We’ve done this dance before (comments #4 and #48 here) The entire point of this piece is that establishing who is “originally responsible” is a distraction, and one used to excuse all manner of shoddy reporting.

    In this case? Yes I would have expected the reporter to question the frankly dubious assertion in the quote. This ain’t quantum physics. It’s a simple epidemiological study and the ground rules for analysing these (a) aren’t hard and (b) have been laid out by folks like Gary Schweitzer FOR YEARS. Even if the scientist was quoted accurately, so what? They screwed up. And in running the quote, so did the reporter. Accountability is not a zero-sum game.

    What you’re saying – that reporters should simply trust everything that’s said to them – is a recipe for disaster. And it’s not journalism; it’s stenography. If anyone actually follows it, they might as well dispense with the science section of their publication and run the RSS feed from Eurekalert.

    The problem with these discussions is not that the problem has been misdiagnosed. It’s that the problem has been identified, but that some reporters have such a warped view of the basic elements of their job that they refuse to be held accountable for their role in screw-ups. And maybe, just maybe, the entire industry would benefit if we cowboyed up and held ourselves to loftier standards, rather than playing a frankly unsavoury game of blame distribution when someone correctly points out a flaw in out work.

  15. Just to be sure: Reading that article (thanks for the lulz btw) I guess that the problem is not reporting the quote, it is believing it and putting it as fact in the title.

    I mean, it’s fine to report that “J.Random Author told us ‘We believe invisible pink unicorns roam in the cosmos’ “, another thing is writing a story entitled “Scientists discover invisible pink unicorns”. Am I right?

  16. That’s only slightly better. Again, if you just uncritically write down what someone says… same problem. You’re not a journalist. You’re a stenographer. And you lead readers into the realm of he-said-she-said reporting, where you can pretty much write whatever the hell you like as long as someone once said it.

    Earlier, you said that journalists could not be held responsible for failure to detect cleverly masked fraud, and there is truth there. But stuff like the Daily Mail example is nothing of the kind. It’s not Marc Hauser or Diederik Stapel. It’s simple stuff that could be identified as exaggeration by anyone with a basic grounding in health reporting. It’s j-school stuff.

    So in your example, I would expect a reporter to note whether J.Random Author had any evidence for their unicorn hypothesis, and to call them out for lack of said evidence if appropriate. Or better yet, ditch the story.

  17. Related discussion from the Truthiness in Social Media Symposium which took place at Harvard University, March 6th 2012:

  18. Am torn between wanting to cheer your great post and quietly weeping that writing it should ever have been necessary in the first place. When I worked as an editor, any reporter who tried to file a story that included quotes lifted from a press release was sent away with a flea in their ear and told to do their own homework. Sadly, the under-resourcing of UK journalism means that doing a decent job of reporting is becoming increasingly difficult.

    I’ve dived in to the Guardian’s discussion about the topic, so won’t repeat myself here, save to say that I think a system of peer review for UK science journalism would be an excellent idea. It would encourage readers to become more demanding consumers of accurate science journalism, and motivate editors/ media shareholders to invest in it more. A number of such initiative exist in other countries, such as Health News Review in the US amd Medien Doktor in Germany.

  19. Completely agree with your post.

    The PCC in the UK should also act faster and have more influence to try and stamp out the mistakes and bad journalism when they occur. This link contains just one example (of many) failings that lead to bad stories being printed in the press.. and nothing being done about it –

  20. dearieme

    Good for you, but don’t let the scientists off the hook. It’s scientists who write abstracts for their papers which are inconsistent with the content of the papers; it’s the same scientists who brief their universities’ spin doctors, and cheerfully appear on the radio or telly to sermonise about their work, explaining the urgent necessity of diverting more money from the taxpayers to themselves.

    P.S. Your remarks would carry more weight with me if you’d chosen an example of bollocks science from the Guardian or Telegraph rather than the Daily Mail.

  21. This is a well written article, driving straight to the point of social decay. I spent twentyfive years of my life as a full time broadcast journalist. Over that period of time I witnessed many of my collegues becoming a little loose in their reporting. It wasn’t that a mistake was made, it became more of a crisis that the mistake was caught. In the case of scientific reporting, the question becomes who will catch an error and who will take responsibility for it and get it fixed?
    Oh, by the way, poor journalistic habits are not limited to the UK’s side of the ocean.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar