8 Comments

Another on-line controversy. (The last I wrote about being the University of California v. Nature Publishing Group, which is on-going.)

Sunday four weeks ago I alerted readers to research that showed that omega 3 oil does not improve cognitive function.*

More recently doctor and well-known anti-’bad science’ writer Ben Goldacre took to task Denis Campbell for writing an inaccurate account of the findings of another research paper on this same topic.

Goldacre‘s criticism has resulted in a minor out-pouring of blog posts.

Before I add a few thoughts to this, here is a more-or-less chronologically-ordered list of posts for those wanting to get ahead of me (or who would rather read the story without my words!):

I have to admit to being tentative about adding my thoughts. The last time I wrote about something along these lines, a few seemed to think I was attacking the person or journalists in general, when I was doing neither!

Readers new to this minor piece of excitement should first read Ben Goldacre’s article. (It is a pity that Campbell’s article is no longer on-line. Skeptic-watch site, Holford Watch, has an article that quotes an an excerpt from it.)

Jeremy Laurance’s account accuses Goldacre of ’pistol-whipping’ Campbell; my reading doesn’t suggest he went quite as far as that! Perhaps Laurance was just trying to spice up his piece with more lively language for interest, and perhaps he is frustrated, but it also strikes me as doing to Goldacre a little what he is asking Goldacre not do.

I agree with some of Fiona Fox’s points, but I don’t see the ’tone’ she refers to, at least not the extent she appears to. Then again, perhaps we’re a more robust lot Down Under?** We have some very roust political journalism in this country, for example… I am under the impression she wants the ’correction’, but without naming names or pointing at the media in general. I’ve no wish to be the judge of this, but wouldn’t this be asking for corrections to be offered without asking for accountability from those that erred?

In any event, my own reading of Goldacre’s article is that his main grief is with inaccurate reporting of science and not citing the research references. (I have recently written about the latter myself, see: To link or not to link: is that the question? and To link or not to link: mainstream media and no links at all.) I’m not going to do a breakdown of his article, enough has been written on that elsewhere and readers can judge for themselves.

A small quibble that I just ’have’ to make about Laurance citing Campbell as having written:

That showed that the fish oil “enhanced the function of those brain regions that are involved in paying attention”, as revealed by a brain scanner.

In addition to the research not being on fish oil, as Goldacre pointed out, fMRI scans show increased activity, not ’enhanced function’. I suspect there may be a confusion (or lack of understanding) of the meaning of the phrase ’functional activation’, which is used in the research paper. (This issue is also explained in the Holford Watch article.)

Goldacre raised the issue of locating the source of the research referred to. As I wrote previously

For formal sources, like research articles or texts, I can see no excuse for not linking to them. None at all. At the very least the sources should be given in plain text in a form that [they] can readily be utilised.

Why the reluctance to cite formal research sources in articles? Here’s one way of looking at it: If the work is so unimportant that there isn’t justification for space to cite the source, why is it being reported?

Ed Yong’s long, reasoned, post is worth reading. Following his remarks about the effort involved I’d add, for what little it’s worth – and my personal thoughts surely aren’t worth much! – that I put considerable effort when I write on this blog. I don’t get paid for writing them either.

Others remark on the issue of fact checking. It reminds me of my comments on an article about epigenetics. What struck me then was that the columnist was writing for himself, rather than presenting experts’ advice on the material. Perhaps a similar issue occurred here? (I don’t have the original to read as it’s no longer on-line.) I’m a little surprised that Fiona Fox didn’t take the opportunity to advertise the UK Science Media Centre’s services.

Here’s a loose thought to close my hopefully-not-too-controversial contribution: if an article presented on-line contains an error, the content should be altered to indicate that error, making it clear that a correction has been made. If the error is too grievous for the content to stand, the content should be replaced with a statement explaining the error. This seems straight-forward to me.

Scientific publications issue errata (or corrigenda) to similar effect.

Like links (references), prompt errata give a measure of credibility, that the publication cares about it’s standards.

Perhaps the traditional media would more convincingly win (on-line) readers’ loyalty by promptly issuing corrections themselves?

To be quite clear, I am asking this as an open question, offering it for comment, not making a rhetoric statement via a question. I’m aware that some publications do something along these lines, but am unsure how commonplace this is.

Footnotes

I still have not gotten around to writing my thoughts on the role of editors in all of this… one day.

* This is the second research article Goldacre refers to in his article, not the first that Campbell wrote his piece about.

** ’Brits’ are renown (or stereotyped, depending on your point of view) for expressing criticism with (sharp) subtlety rather than blunter statements.


Other articles on Code for life:

What is your relationship with your research notebook? (Researchers have intimate relationships with their notebooks)

Autism genetics, how do you copy? (A ’lite’ summary of the recent Nature paper on CNVs in autistic people)

To link or not to link: is that the question? (Is linking with articles really a style question or a technological problem?)

New Ministry of Science and Innovation for New Zealand (Join the discussion about governance of science)