How Aric Sigman distorts the scientific evidence to mislead you.

February 24th, 2009 by Ben Goldacre in aric sigman, bad science, mail, onanism, references | 153 Comments »

I was on newsnight a second ago, debating the rather indulgent claims of Baroness Professor Susan Greenfield and Dr Aric Sigman about Facebook and Twitter. It’s 40 minutes in to the show, which can be seen here as a wmv/rm file or here on iPlayer or here:

I promised references. These can be found below.

Professor Susan Greenfield is the head of the Royal Institution and the person behind the Daily Mail headline “Social websites harm children’s brains: Chilling warning to parents from top neuroscientist”, which has spread around the world (like the last time she said it, and the time before that).

It is my view that Professor Greenfield has been abusing her position as a professor, and head of the Royal Institution, for many years now, using these roles to give weight to her speculations and prejudices in a way that is entirely inappropriate. Sometimes it’s cannabis.

We are all free to have fanciful ideas. Professor Greenfield’s stated aim, however, is to improve the public’s understanding of science: and yet repeatedly she appears in the media making wild headline-grabbing claims, without evidence, all the while telling us repeatedly that she is a scientist. By doing this, the head of the RI grossly misrepresents what it is that scientists do, and indeed the whole notion of what it means to have empirical evidence for a claim. It makes me quite sad, when the public’s understanding of science is in such a terrible state, that this is one of our most prominent and well funded champions.

Then there was Dr Aric Sigman. He is the man behind the “Facebook causes cancer” story in the Daily Mail, and many other similar stories over the years (as part of the Daily Mail’s ongoing oncological ontology project). His article can be read in full online here as a PDF. I explained that he had cherry picked the evidence in his rather fanciful essay, selectively only mentioning the evidence that supports his case, and ignoring the evidence that goes against it.

Cherry picking is a common crime in the world of pseudoscience – whether it is big pharma or everyday cranks – and to me it is a serious crime against science, because by selectively quoting evidence, you can make almost anything seem either dangerous, or beneficial.

Dr Sigman’s case is that social networking leads to loneliness, and loneliness leads to biological harm (he doesn’t mention cancer specifically, incidentally). I didn’t get near the second half of his argument, though: because he was so spectacularly misleading on the first that it became irrelevant.

I claim no expertise on the question of whether social networking and internet use is linked to loneliness. I merely have a basic ability to use searchable databases of academic evidence, like anybody else. If you go to PubMed and type in:

loneliness [ti] AND internet

you will get 12 results.

Many of them do not support Dr Sigman’s theory. These are the ones he completely ignores.

For example:

1. Caplan SE published a paper in 2007 entitled: “Relations among loneliness, social anxiety, and problematic Internet use.” Dr Sigman did not quote this paper in his article. Why not? “The results support the hypothesis that the relationship between loneliness and preference for online social interaction is spurious.”

2. Sum et al published a paper in 2008 with the title: “Internet use and loneliness in older adults“. Dr Sigman chose not to quote this paper. Why not? I don’t know, although it does contain the line “greater use of the Internet as a communication tool was associated with a lower level of social loneliness.”

3. Subrahmanyam et al published a paper in 2007 called “Adolescents on the net: Internet use and well-being.” It features the line “loneliness was not related to the total time spent online, nor to the time spent on e-mail”. Dr Sigman ignored it.

And so on.

I am not claiming to give you a formal, balanced, systematic review in these examples, I am simply demonstrating to you the way that Dr Sigman has ignored inconvenient evidence, in order to build his case.

Was this an oversight? Were these papers hard to find? I think not. And Vaughan at the ever-brilliant Mindhacks.com (surely the best popular psychology content anywhere, not just on the net) found some even more damning evidence.

He points out the Dr Sigman quoted a 1998 paper called “The Internet Paradox”. This paper did indeed find a (weak) relationship between internet use and depression, loneliness, etc. This was 1998, at the very dawn of widespread use of the web, but more importantly, the very same authors went back and looked at the very same families, and found that the effect had disappeared. That seems relevant to me, especially if you’re going to quote the 1998 results, Dr Sigman?

You can read the paper in full online as a pdf. It says “This sample generally experienced positive effects of using the Internet on communication, social involvement, and well-being.”

There is no excuse for not knowing about this finding. Type the internet paradox into Google. Go on, do it:

www.google.co.uk/search?q=the+internet+paradox

“The Internet Paradox Revisited”, the paper Dr Sigman ignored, is the second result.

Dr Sigman claimed, on and off camera, that his was an opinion piece and so it is acceptable to quote only half the evidence. This is ricockulous, this is not how the paper was presented in the media, it is not what people were expecting, and it’s not what I’d expect from any opinion piece in anything approaching a scientific journal. In fact, this way madness lies: comment is free, but facts are sacred. If you cherry pick your evidence, you can make a very good case that all swans are black. This would not represent a useful argument.

Excitingly, Dr Sigman then went on to claim that his article “Well connected? The biological implications of social networking” was not in fact about social networking.

It is quite hard to have a meaningful discussion with someone like this.

Update:

A friend also just sent me this excellent review from 2009 on facebook and cancer. We might forgive Sigman for not referencing it but it’s a good summary for those who are interested.

www.cam-ascor.nl/images/documents/2009_Valkenburg__Peter_Current_Directions.pdf


++++++++++++++++++++++++++++++++++++++++++
If you like what I do, and you want me to do more, you can: buy my books Bad Science and Bad Pharma, give them to your friends, put them on your reading list, employ me to do a talk, or tweet this article to your friends. Thanks! ++++++++++++++++++++++++++++++++++++++++++

153 Responses



  1. danman said,

    February 29, 2012 at 1:11 am

    Almost enuff said really. She goes on and on and on and with unfailing regularity and has that uncanny knack of talking, nay, spouting, way beyond what data she ever had. She’s been doing it ever since she was a PhD student and never ceases to disappoint. Oxford does have a bit of a habit of indulging this kind of crap and I suppose that the RI is a quasi Oxford college in all but name. But really, Professor of P.U.S? It is an outrageous farce and the public really do deserve better. The BBC should certainly stop indulging her fantasies but then they recently tried to install Alan Titmarsh as a David Attenborough substitute to talk about the Nature of Britain. There’s no hope really. daily car insurance

  2. kendersrule said,

    May 2, 2012 at 12:42 pm

    Sigman’s article has vanished from the land of PDF!

  3. kendersrule said,

    May 2, 2012 at 12:44 pm

    twitter friends are generally people I’ve YET to meet. Not: have not met/will never meet. Big difference.