'More people living in Britain'

An unintentionally ambiguous start to a story in today's Metro (under the headline 'We are, in fact, proud to be British'):

More people living in Britain see themselves as British first and foremost - whatever their background.

Is that 'more' as in the majority, or 'more' as in more than before?

The former, according to the rest of the story, although I would have assumed the latter.

Shock horror: DJs occasionally mention alcohol

Today's edition of the free paper Metro contains a story entitled 'DJs 'encouraging drinking'' which states:

Radio presenters widely encourage drinking, a study warns today. DJs use language that celebrates excessive drinking with research into 1,200 hours of radio output showing 703 references to alcohol.

So - 1,200 hours of radio and 703 references to alcohol. Maths isn't my strong point, but I make that one reference to alcohol about every hour and three quarters, on average. And that's any reference to alcohol, not necessarily celebrations of "excessive drinking". Hardly a shocking statistic.

To be fair, the online version of the story includes far more detail and is the better for it. However it baffles me for a different reason, stating as it does:

Research into 1,200 hours of radio output uncovered 703 extracts containing references to alcohol.

Of these, 179 involved comments made by presenters and another 45 were comments initiated by co-presenters, studio guests and members of the audience.

So who or what made the remaining 479 references to alcohol if not presenters, co-presenters, studio guests and audience members? Jingles, maybe?

It's worth noting that the study was funded by "the Department of Health and the Home Office as part of the Know Your Limits campaign". So not biased at all, then.

And sorry if this blog post is slightly less clear than it might be. I've been conducting some research of my own this evening...

Teachers, doctors, pupils and statistics

I love the ambiguity in this clipping taken from Personnel Today magazine (and sorry it's wonky, that's just my usual inability to cut and/or scan things straight):


Does this mean that 12% of teachers who have been attacked by a pupil subsequently needed to visit a doctor, or that 12% of teachers have been attacked by a pupil and subsequently needed to visit a doctor?

If you can't see or don't care about the difference, you might not want to become a sub...

Half the country hasn't an NHS dentist

Gareth sent in this scan of a recent Daily Mail front page with a question about the sentence construction used in the headline, but I'd been meaning to blog about this story for a different reason.


If you can't see the scan, I should tell you that the headline reads "Half the country hasn't an NHS dentist"; the web version of the same story explicitly states what the headline implies, namely that "Half the country can't get an NHS dentist". And perhaps for some of our foreign readers I should explain that the NHS is the National Health Service – the role of which is to provide health care, often free, to anyone normally resident in England (Scotland and Wales apparently have their own Health Services).

In contrast to the headline, the story itself states only that:

In total, 23,161,368 people in England - almost half the population - received no dental care on the Health Service in the two years up to last September.

To my mind, there's a great deal of difference between someone being unable to get an NHS dentist and someone not receiving dental care from an NHS dentist. After all, there are many reasons why people might not go to an NHS dentist in any given period: they might choose to have private dental care; they might have a phobia of dentists; they might not think about going to the dentist at all.

So to say that half the country can't get an NHS dentist simply because they don't go to an NHS dentist is highly misleading. It's part of a worrying trend of NHS-bashing I've noticed in the media – other popular soft targets include the BBC and the public transport network. Of course, all of these have their failings but there's no need to invent new failings by deliberately misinterpreting statistics.

In my own experience, which probably isn't representative of anything, I've had no problem finding an NHS dentist willing to take on new patients. It does seem usual to have to pay for a private hygienist though...

Metro: How many white people in Britain?

It's dodgy statistics time again, today courtesy of free paper Metro.

In the midst of an feature about 'the state of the nation', based on a National Statisticians article on society, Metro makes the following claim:

Percentage of UK population classified as white
2007: 90%
(England 65%)

Um, given that England makes up about 50 million of the UK's 60 million-ish population, this just doesn't add up.

If 90% of the UK's 60 million population is white, that's 54 million of them who are white and 6 million not. Even if every one of those non-white folk lives in England (which isn't the case), that still gives England a white population of 44 million, or just over 80%.

So either the figure of 90% is wrong, or the figure of 65% is wrong. Or possibly both. When I spot something like this in a feature, it immediately makes me mistrust the entire thing and I usually just stop reading.

Just a thought: any chance that figure of 65% was supposed to refer to London?

Tearing things out of papers
is not my forte...

Nursing Times: sexual relationship poll

It's misleading statistics time again today. A Nursing Times poll reports that:

Up to 16% [of nurses] said they knew of a colleague who had started a sexual relationship with a patient during the time that person was in their care

This was reported in one of the free London papers as "one in six nurses...". Somehow that makes it sound even worse (never mind the fact that one in six is actually nearer 17%).

But however you present it, the figure is actually pretty low. After all, the poll isn't saying that one in six nurses has started a sexual relationship with a patient, but that one in six nurses knows of a colleague who has. And how many colleagues does a nurse have – 10? 20? 50? Over the course of a career, that figure must easily run into hundreds (after all, the poll doesn't specify 'current colleague'). The chances are that if it was at all common for nurses to start sexual relationships with patients, almost every nurse would know at least one colleague who had done so.

Or looking at it another way, the vast majority of nurses have never known a single colleague who started a sexual relationship with a patient. Conclusion: either it is very uncommon for nurses to have sex with their patients, or they just do it on the sly...