Tuesday, 18 May 2010

Lies, damned lies, and headlines

(Updated, to add a headline. Oh, the irony).

This is, maybe, a story of how crap science reporting starts with the headline.
Actually, it hardly ever does: The story arrives, gets written by a journalist, and the headline is probably the last piece of the puzzle, going through sub-editors and final print editors before being set in stone (or, well, type. Or hypertext).
But how does the innocent reader arrive at a story? Via the headline.
Which makes it important.
The story is this:
A long-term (10 year retrospective) and large (nearly 13,000 users) study of mobile phone usage, investigating links with cancer (yes, another one), showed no definite results either way save for the top 10% of users, who had an up to 40% greater risk of developing a brain tumour, based upon self-reported usage patterns, which are methodologically one of the weakest ways of assessing actual usage and introduces problems the study authors are not shy to point out.
These are the headlines from four media outlets:
ABC News (copy sourced from Reuters, headline by ABC): Mobile phone-cancer link study inconclusive
Courier Mail (based upon the Australian story): Brain cancer link to mobile phones
Did you experience a sinking feeling as you read down? You should have.
The only dedicated science outlet uses a catchy, populist approach to combine the main finding (no real link) and the main problem (not reliable enough). The ABC goes for a direct and accurate but hardly exciting approach. The Australian pulls out the most news-worthy finding to focus upon and ignored 90% of users, and the Courier Mail just throws everything to the wind and goes with a misleading title for a report written to emphasise just one finding and foment debate.
It's depressing, it really is.
The media has an important, even central, frequently even solitary, part to play in informing the public and part of that is, some of us would like to believe, a responsibility to be accurate and not monger fear or incite hysteria.
I would also like to think that if we assume the public is intelligent enough to read an article saying "probably not, but there may be a risk if you do something a lot", we may be astonished to find out they are.
And then there's historical perspective.
The fifth paragraph in the Reuters/ABC copy is this:
Years of research have failed to establish a connection.
And do you really think people haven't tried, really hard? I've been using a mobile for ten years now and I remember the debate happening years before I started.
Basic Principle of Coping With Reality: It's complicated, it's not your only risk factor, and if it was really a problem we'd probably have found something by now.
We'd also have found a link to testicular and uterine cancer from all the phones carried on or about hips, I'm thinking.
Consider as well just how any radio waves saturate the air from TV, radio and other people's mobiles, as well as short-wave, long-wave and GPS radio signals. If you're going to worry about electromagnetic radiation, a mobile is the least of your problems.
As a bonus: Look up the difference between ionising and non-ionising radiation. I leave that as an exercise for the reader.

No comments:

Search This Blog