Tuesday, July 03, 2007

Never Believe What You Read In The Papers

I'm writing an article for a client that's partly about the growing use of internet banking in the UK.

When I asked the client if he had any figures illustrating the growth of online banking, he directed me to two recentish press releases.

The first press release was issued by Apacs on the 28th December last year, and says that 'over a third' of UK adults now bank online, with the 16-24 age bracket growing fastest.

The second release was issued 25 days later by Lloyds TSB, and says that 68% of UK adults now bank online, with the over-50 age bracket growing fastest.

Both stories were reported faithfully by the financial newswire Finextra, with no acknowledgement of - let alone attempt made to investigate - the gaping discrepancies evident therein.

It annoys me intensely that a lot of 'journalism' is actually just re-printing press releases about spurious surveys that miraculously support whatever it is the companies who wrote them are selling.

Next time you see some survey data in the paper, take care to note who sponsored or carried out the survey. If it's a company promoting an associated product, pay close attention to the small print, especially about how many and what sort of people were surveyed and what they were asked. Nine times out of ten* it will turn out to be statistically invalid, spurious bollocks.

I wish more journalists had the presence of mind to do the same.

* Hee.


Fidel said...

76% of female bloggers are single women in their 30's.

42% of male bloggers still live with their mothers.

48% of politicians are left handed (Morganatic) against a national background statistic of 17%.

11% of adult-male internet users remain virgins

Sylvia said...

Since the papers don't seem to report anything anymore except the opinions of journalists and their mates, I simply don't bother with most of the news sections of papers these days.

I go straight for the cartoons!

chuffy! said...

86% of financial journalists like an easy life.

HIJACK...blog live and unread here. Slightly different persona, similar schtick.

James said...

That's scary although not totally surprising.

I suppose you could argue that modern media may help eradicate this. By this I mean media where there is interaction with the reader/viewer means that a dodgy basis to a piece can be highlighted, with the author losing face.

I thought about leaving some statistical jokes but I think I'm in the 2% of the population who lacks that ability.

ScroobiousScrivener said...

Nine out of ten journalists are idiots.

No, that's not fair. But many of this sort of story - press release, as you say - are given to the junior reporters to polish up, and they simply don't have the experience to apply brain in the right ways, especially when it comes to numbers.

True story: I was once subbing a story that claimed "dramatically more X than Y [like to practise Z]". Further down, some actual numbers materialised. 49% of X, it turned out; and "half" of Y. I challenged the writer on it, and he said oh, er, yes; well, it was actually 53% of Y, but he'd rounded it out to sound better. Fine, but even so, that's not dramatically more of anything. It's not, actually, even a story.

(Of course this shouldn't have made it past the news editor. Bad day all round.)

Tim Footman said...

What annoys me most is that hacks who rewrite corporate press releases as news items regard themselves as being somehow morally superior to people who are prepared to rewrite a phone conversation from Jodie Marsh's PR underling as a news item.

I once pinned down a hack who'd written an auto-related story that mentioned "enhanced triple-torque pokepower ramshafts" or some such nonsense.

"What does that mean?" I asked.

"I have no idea," he said. "But I guessed that readers who are interested in cars would understand it."

90% :-) of news stories can be replaced with slabs of 36-point saying "CAPITALISM IS GOOD. LOVE IT"

Anonymous said...

Time I finished working last night?

After 10.

No. of words I need to write by deadline at midday today?

About 4,000.

Will there be some press release re-hashing?

Fuck, yeah!

It's not always down to laziness...


patroclus said...

Hello all, yes, sorry, bit ranty yesterday. Survey press releases are a particular annoyance of mine. God knows I've written enough of them, and I know how contrived they can be and how poor the statistical sample usually is.

If there was any justice in the world, news editors would weed the obviously spurious ones out and throw them in the bin, but I understand there are column inches to fill - and who cares if they're filled with made-up stuff from PRs? The whole of Cosmo, Grazia, Observer Woman and their ilk is filled with made-up stuff from fashion and beauty PRs; that's how women grow up believing that the way they look and dress is more important than anything else.

I'll stop ranting now before I turn into Charles Arthur.

PS Hello S! That sounds horrific - are you producing the whole paper?

Sean McManus said...

Sue Townsend cites the statistic that 8.3% of disciples betrayed Jesus.

There was a report last year that said xx% of people don't trust government statistics, which I thought was nicely circular.

GreatSheElephant said...

Methinks you mistake the purpose of trade organs like Finextra. What Tim said.