Archive for category Damned lies and statistics

Hangovers, statistics and a dodgy hookup

It’s one of the oldest clichés in the book. You go to a party, get completely hammered, and wake up in bed with a dodgy PR firm.

Today’s ill-advised hookup is a threesome between The Express, The Mirror and a non-alcoholic drinks company called Sweet Lady Beverages, who claim that “the average Briton will spend five years of their life with a hangover“.

Before we look at the article itself, a quick sanity check. Life expectancy in the UK is roughly 80 years, and it’s unlikely people are going to experience hangovers before the age of about 15 or so. So, at maximum, that gives the average Brit about 65 drinking years. If the Express‘s statistics are true, we spend 8% of our adult lives hung over – we would spend more time hungover than we would eating. It’s amazing anyone gets anything done.

The article goes on to say that:

[Britons] will suffer the ill effects for a whole day – usually a Sunday – at least once a week between the ages of 21 and 38.

Bear in mind that this an average. According to Sweet Lady Beverages, the average person is hung over every week until the age of 40, and those hangovers last all day. That sounds a tiny bit excessive. After all, one – much more scientific – study found that having even just one hangover per month over an extended period is linked to a major increase (around 2.36 times) in heart attack risk.* And yet somehow, we’re not dropping like flies.

As far as I can tell – there’s no information about this survey available on the web outside these two articles – Sweet Lady Beverages simply asked visitors to its site to answer some questions about hangovers. There’s no published methodology; in other words, they don’t say what questions were asked or what precautions they made to make sure they had a fair sample.

For instance, they could have asked

It would certainly explain the odd results they got.

The Sweet Lady Beverage company is quoted by the Express as saying

The message we can take from this is simple – by reducing our alcohol intake we can reduce the amount of time feeling wretched.

Oddly on-message for a company selling alcohol-free drinks, wouldn’t you say?

* I can’t find many good scientific studies of hangovers. A lot of them are rather hamstrung by the fact that surveys usually take place in university, and therefore involve university students – not very representative of the drinking habits of the wider population! Nevertheless, this paper suggests that only 15% of the population have more than hangover per month.

Edit: The Daily Mail has now picked up the story too.

, ,

Leave a comment

Evidence? Where Cameron’s going, he doesn’t need evidence!

David Cameron is going to give a speech today with Lord Reid today about the AV referendum. In it, he will say:

“Too often debates about AV are less like political arguments, and more like scientific discussions, where people get lost in a language of proportionality and preferences, probabilities and possibilities.

“Of course, some of these things are important. But for me, politics shouldn’t be some mind-bending exercise. It’s about what you feel in your gut – about the values you hold dear and the beliefs you instinctively have. And I just feel it, in my gut, that AV is wrong.”

Just a reminder, this came from the Prime Minister of the United Kingdom, not a Richard Littlejohn column or some particularly stupid Comment is Free post. The man who directs a nation of 62 million people and controls the sixth largest economy in the world doesn’t think politics should be about thinking and weighing up different options. Oh no, that’s just “some mind-bending exercise”! No, politics is about what you feel in your gut* and your blind instincts.

Voting systems are, at heart, all about mathematics – each voting system is just a different way of counting people’s opinions. You cannot discuss any voting system without taking into account the way it behaves mathematically. The language of “proportionality and preferences, probabilities and possibilities” is not swamping the debate, it is the debate.

Imagine if this was any other debate. Imagine if David Cameron said we should ignore the clinical trials when deciding whether to fund a particular drug, or the climate models when considering pollution controls, on the grounds that all this scientific evidence was “mind-bending” and got in the way of his “instinctive beliefs”. No-one would think that was an appropriate way for an elected official to make decisions.

If you push the evidence out of the debate, all you’re left with is empty sloganising, blatant untruths and tribal party politics. Without a proper debate on the pros and cons of each voting system, the AV referendum just becomes a Cameron/Clegg popularity contest. That’s a terrible way to decide an issue that will shape government in the UK for decades to come.

* Most of what’s in your gut is digested food, so perhaps Cameron is just saying that politics should be full of shit?

7 Comments

Another day, another stupid asteroid scare

The road I live on has a railway running across the end of it. Every day, hundreds of tons of metal speeds along the line just a hundred metres or so from my house. Yet I don’t live in fear of waking up one morning and finding a train’s crashed into my house, because of course the trains are restricted to the railway tracks.

Space is much the same. Asteroids are whizzing around over our heads every day, but they follow precisely defined orbits through the sky. An asteroid passing close to the Earth is no more a “near miss” than a train passing my house without hitting it is a “lucky escape”.

It’s a simple enough idea, you’d think, and yet…

“Phew, that was a near miss: Cigar-shaped asteroid stronger than ’15 atomic bombs’ whizzes by earth”

Thankfully, the 50m long rock that could have destroyed a small country went barely noticed as it passed earth at a distance of some 2,085,321 miles.

Yes, the asteroid 2011 GP59 could have destroyed a small country. If it was two million miles closer.

Once again, this article has been taken from the Australian news site news.com.au, who seem to have a thing for scaremongering stories about space; they also started the rumour that Betelgeuese would go supernova in 2012 and gave credence to the shameful “supermoon” story. At least the Mail‘s headline is less awful than news.com.au’s, who’ve gone with “Scientists find asteroid with potential power of 15 atomic bombs. Heading this way. Tonight.” which surely has to rival shouting “fire” in a crowded theatre in terms of unethical scaremongering stupidity.

Incidentally, the claim that it’s “as powerful as 15 atomic bombs” doesn’t come from any scientific authority. It comes from the news.com.au journalist – who doesn’t appear to be a science journalist at all, but a technology journalist – digging up an old New Scientist article about an asteroid that exploded with the energy of three nuclear bombs (three of the very small Hiroshima bombs, I should point out, not a modern nuclear bomb), and then scaling it up. This is a stupid calculation for a number of reasons:

  1. You can’t just say “this asteroid is 10 metres long, this asteroid is 50 metres long, therefore it’s 5 times bigger”. It’s the volume which is important – the length times the width times the height. Assuming the asteroid is 5 times bigger in each direction, then it’s 5 x 5 x 5 times bigger, which is 125 times the size. If the journalist hadn’t cocked up his maths, he could have made this asteroid sound EVEN SCARIER. Except…
  2. The amount of energy an asteroid has depends on its speed. A fast moving asteroid carries far more energy than a slow moving one, and a small increase in speed causes a much larger increase in energy.** The gravitational pull of the Earth as the asteroid approaches plays a large role in determining its speed, so the energy it would have would depend on the route it took to Earth. Since this asteroid is not heading for Earth, it’s meaningless to ask how much energy it would have if it hit Earth.
  3. It also depends what the asteroid is made of. Most asteroids are made of dust and ice, and burn up harmlessly in the atmosphere. A few – maybe one per year – explode high up in the atmosphere. And a very, very few – mostly large metallic asteroids that don’t burn as well –  hit the ground. Again, we don’t know what this asteroid is made of.
  4. THE ASTEROID IS TWO MILLION MILES AWAY AND WILL NOT HIT EARTH AT ANY TIME IN THE FORESEEABLE FUTURE.

At any rate, there are literally thousands of asteroids this big – or indeed much bigger – rattling around near-Earth space, and there must be thousands more we haven’t detected yet. It’s worth being sensibly worried about the risk of currently undiscovered asteroid hitting us, but getting worked up about an asteroid that we know can’t hit us is just stupid.

Now, if you’ll excuse me, I’m off to dig a train-proof bunker in my garden.

* Psst, Daily Mail Reporter. “Earth” has a capital “e”.

** In fact, energy is proportional to speed squared – if you double your speed, your energy goes up fourfold.

,

1 Comment

Channel 5 and the Express – some unsurprising statistics

Inspired by my last post, I’ve decided to have a quick look at the Express‘s TV review archive.

Channel 5 was bought by Richard Desmond, owner of Express Newspapers, on 23 July 2010.

So far this year, a Channel 5 show has been a “Pick of the Day” 38 times (warning: all links are direct to the Express site)

  1. Law and Order, 18th Feb
  2. OK! TV, 17th Feb
  3. Starlight: For the Children, 16th Feb
  4. Stansted: The Inside Story, 15th Feb
  5. OK! TV, 14th Feb
  6. Law and Order, 11th Feb
  7. Julius Caesar: Rome Unwrapped, 10th Feb
  8. NCIS, 9th Feb
  9. Stansted: The Inside Story, 8th Feb
  10. The Vanessa Show, 7th Feb
  11. Royal Navy Caribbean Patrol, 7th Feb
  12. Ice Road Truckers, 4th Feb
  13. Secrets of the Vanishing Sphinx, 3rd Feb
  14. The Punisher, 3rd Feb
  15. NCIS, 2nd Feb
  16. CSI, 1st Feb
  17. Home and Away, 31st Jan
  18. Britain’s Secret Schindler: Revealed, 27th Jan
  19. Justin Lee Collins: Turning Japanese, 27th Jan
  20. Cowboy Builders, 26th Jan
  21. CSI, 25th Jan
  22. Neighbours, 24th Jan
  23. Secrets of the Blitz: Revealed, 20th Jan
  24. Cowboy Builders, 19th Jan
  25. CSI, 18th Jan
  26. The Wright Stuff/The Vanessa Show, 17th Jan (two separate shows both on Channel 5)
  27. Pirates of the Caribbean: The True Story, 14th Jan
  28. Ripped from the Cockpit: BA Flight of Terror, 13th Jan
  29. Neighbours, 12th Jan
  30. Jack the Ripper: The Definitive Story, 11th Jan
  31. CSI/Taggart, 11th Jan (CSI is on Channel 5, Taggart is on ITV)
  32. The Vanessa Show, 10th Jan
  33. How Do They Do It?, 10th Jan
  34. Titanic: The True Story, 7th Jan
  35. Highland Emergency, 5th Jan
  36. Rhino In My House, 5th Jan
  37. How Do They Do It?, 4th Jan
  38. Goering’s Last Secret Revealed, 4th Jan

There have been 35 weekdays so far this year. On average, that means a Channel 5 show is a pick of the day slightly more than once a day – 1.09 times per day to be precise.

Over this time last year, a Channel 5 show (back then, the channel was called “FIVE”) was a pick of the day 3 times:

  1. Paul Merton in Europe, 15th Feb
  2. The Gadget Show, 8th Feb
  3. The Gadget Show, 1st Feb

That works out at roughly one pick every twelve days, or 0.085 picks per day.

CSI and Law and Order were both on FIVE at the time, but funnily enough, neither of them were mentioned at all over this period. CSI was hardly ever an Express pick of the day before the takeover – from the looks of it, it got four mentions between 2007 and 2009, or about twice a year. Since October 2010, however, it’s been pick of the day 10 times in just 16 weeks.

I know it’s hardly surprising that the Express and Channel 5 are a bit chummy, but that doesn’t make it any less of a shock when you see it in terms of raw numbers.

Edit: Thanks to Lazer Guided Melody in the comments for pointing out that the statistics need more context. Here’s a very quick count of the number of recommendations over the same period for each of the major channels:

BBC1: This year, 25. Last year, 48
BBC2: This year, 21. Last year, 13
ITV1: This year, 31. Last year, 42
Channel 4: This year, 8. Last year, 37
Channel 5: This year, 38. Last year, 3

4 Comments

A survey? How romantic!

Blah blah romance is dead, say The Express and The Mirror. Why?

Only nine per cent of those surveyed had ever sent a letter – and most of those were over 50 – while more than two-thirds prefer to say “I love you” by text.

An additional 24 per cent would rather send an email to express their feelings, while 14 per cent said they would post a message on their lover’s Facebook wall.

In further shocking news, very few couples surveyed courted via telegraph, and fewer still stood on hilltops furiously waving sweet nothings in semaphore. The Express quotes relationship expert Jo Barnett saying this shows that “[people] want an instant relationship with instant physical contact, they feel they’ve not got enough time to romance their partner” rather than the more immediately obvious conclusion that people are just sending fewer letters in general.

If the love letter is dead, then it seems male chivalry is also on its last legs as just four per cent of men said they would send flowers to their partner’s place of work. Only five per cent stand up when their partner stands from the dinner table, while only 12 per cent have booked a surprise weekend away.

This is the same rubbish that comes out every few months; pearl-clutching panic about how “chivalry is dead” (a phrase which appears 49 times on The Mail‘s site, incidentally) when it would be more accurate to say “society’s norms about how men should treat women are (slowly) changing to be less infantilising” (a less snappy headline, I’ll admit). After all, I’m pretty sure if I stood every time my girlfriend left the table, she wouldn’t find that chivalrous but instead rather creepy.

Anyway, I’m sure there was a perfectly valid, academic reason this survey was done, right?

The study was to mark the DVD release of romantic comedy Going the Distance.

Going the Distance is released on Blu-ray Triple Play and DVD today, from Warner Home Video.

Oh.

,

7 Comments

Short and royally silly statistics

Being a monarch is four times more dangerous than being a soldier fighting on the front line! Charles and William are doomed!

Well… if you look at every European king and queen since the year 600, and assume that the regicide rate today is exactly the same as it was a thousand years ago, that is…

The Guardian‘s coverage is a bit better – at least they seem to know the difference between past and present!

2 Comments

News of the World shows us how not to use statistics

Thank god for churnalism. The News of the World yesterday published an article claiming to have found the most “workshy” neighbourhood in Britain, but of course that ended up locked behind its paywall. Luckily, The Daily Express and The Daily Mail have both churned out articles based on NotW‘s, so I don’t have to pay Murdoch to read it.

The community that they claim is most “workshy” is a small area called Cottsmeadow Estate in Birmingham. Before I even go into the statistics of this, I’d like to point out that having a lot of people on benefits does not mean an area is “workshy”. Perhaps it’s an area where a major employer recently went bust. Perhaps it’s an area with a lot of affordable, accessible housing perfect for disabled people. Or perhaps, as seems to be the case here, it’s an especially deprived area, which has been very hard hit by the recession. After all, the majority of people in the area receiving benefits are getting Jobseekers Allowance (and quite a few more people are receiving income support, which means they work part-time).

Anyway, the newspapers claim that 106 people of working age live on the estate, of whom 105 are on benefits. Population statistics for individual “census output areas” are only available by request, annoyingly, so I’ll have to take that on faith for now. However, the population data is just an estimate, not a robust census, and when you’re dealing with areas as small as 100 people (out of a population of 60 million), you’re bound to have quite a bit of error in there.

The newspapers claim to have tracked down the lone worker – the mind boggles over how they could possibly have gone about this (did they go from door to door asking people “do you have a job?”), especially since the data in question dates from June of last year – in employment terms, that’s rather stale. Now, if they had found the only person on the estate who wasn’t receiving some sort of benefit, they’d almost certainly have breached the Data Protection Act – giving the name of the only person who does not receive benefits is, in effect, exactly the same as revealing everyone else does. This is precisely why the DWP anonymises their data – they randomly round each figure up or down to a multiple of 5, so you can’t work out who is or isn’t on benefits by taking advantage of small numbers.

In this case, it looks like they’ve probably underestimated the population of the area. After all, according to the statistics, three months earlier there were 110 people receiving benefits in the area (code 00CNGP0059) – more than the estimated population! This seems to be the only reason to focus on such ridiculously small areas. The data is divided into “census output areas” – the smallest division that the Office for National Statistics uses, and therefore most error prone too. Looking at the ward Cottsmeadow Estate is in, Washwood Heath, there appears to be about 4,950 people receiving benefits out of a working age population of around 15,000. This is a sample almost 150 times larger than just Cottsmeadow Estate and a much fairer way to gauge the number of people receiving benefits in the area.

These articles, had they been written properly, could have carried an important message – some areas are more deprived than others, and we need to make sure that everyone has access to work. The way the Mail and the Express (and presumably NotW, but alas I don’t have a copy of the article) cover it however completely destroys any attempt at nuance, tarring whole neighbourhoods as being full of “workshy” “scroungers”, regardless of what the statistics and basic common sense say.

, ,

2 Comments

Another anti-contraceptive scare story

Happy new year to everyone!

I don’t normally mention TV news here, but they can slip up too. Channel 4 News yesterday ran a big, scaremongering piece about one simple statistic: 584 people with contraceptive implants became pregnant.

This might be newsworthy, except Channel 4 forgot to mention two rather important things, subsequently picked up on by the BBC.

First of all, the data in question covers 11 years, not just one year.

Secondly, over that time, the implant has been used by around 1.4 million women.

Now fair enough, presumably not everyone who got pregnant after using Implanon reported it, and contraceptive failure is always regrettable. 584 pregnancies among 1.4 million users however means that the implant did not fail in 99.95% of patients. That is very, very reliable in medical terms.

For comparison, vasectomy is 99.9% effective, an IUD is 99.8% effective, the pill is 99.7% effective (when taken properly; people missing doses means that in real life, it’s only 92% effective on average) and condoms are 98% effective (again, when used properly).

It’s always good to make sure people are completely aware of the relative risks of any type of contraceptive (and indeed any medicine), but using these 584 pregnancies as a sign that there’s something wrong with the implant, without any kind of context or an explanation, isn’t going to do this. All it will do is scare people – as Channel 4 have now realised. They’ve since released another article, “Implanon implant: what to do if you’re worried“, which explains:

You do not need to speak to your doctor unless you are very worried and need to have your mind put at rest.

As long as you can feel the implant, there is no cause for concern. The implant is still a very popular, safe and reliable method of contraception.

No method is 100 per cent effective but only a tiny number of women using the implant have got pregnant.

Good advice, but they should have put that in the actual article yesterday.

, ,

4 Comments

The Delingpole Escape Plan

(Why yes, I did write this post entirely for the pun.)

The Eco-fascists have taken over! So says James Delingpole, anyway. His proof?

Well, in October, Brussels Airport won an award for reducing its carbon emissions. A month later, it snowed.

This isn’t a new thing. It has been going on for years, since at least the 1992 Rio Earth Summit when Maurice Strong laid down the ground rules for the eco-fascist takeover of the world. It’s bit like that classic 2000AD “Future Shocks” story where the aliens that have invaded our planet unbeknownst to us turn out to be those innocuous-looking wire coat hangers we have in our cupboards. The battle for our freedom is already all but lost – and the stupid thing is, most of us didn’t even know we were fighting one.

So now – not in some imagined, paranoid fantasist’s future, but NOW – we live in a world where an airport is encouraged to place a higher priority on reducing its notional production of a harmless trace gas than it does in making provision for aeroplanes to be able to take off and land in inclement weather.

You may have noticed it was not snowing in October, and that it’s perfectly possible for an airport to offset its carbon emissions and clear away snow, and that the two events were completely unrelated. Shush!

Delingpole then moves onto the Met Office. The Met Office’s seasonal forecasts are always vague – as winter approaches, they generally make them more accurate. True enough, in the last few years, they’ve generally predicted that a warm winter was more likely than a cold one, and had to revise that as winter approached. Whether or not the Met Office was right has nothing to do with climate change however. A year is much too short to see global warming happen.

Delingpole quotes a Department for Transport report (pdf) which, he claims, shows the Met Office predicted there was only a 1 in 20 chance that this winter would be severe (which, of course, does not mean the same thing as simply “cold”). In fact, this is taken out of context – the report simply points out (p. 88-9) that in general, one winter in twenty is severe, and the fact that the last two years were severe does not necessarily mean they are clustering. The report after all was issued in July, long before the Met Office could say with any certainty what the weather would do.

Incidentally, the Met Office didn’t predict this winter would be mild. One of their computers predicted the winter would be mild, based on a limited set of data, but as the Met Office said at the time:

“This is not an official forecast, it’s data that would form part of a longer term prediction.

“If you look at the whole picture across north west Europe, there’s a higher chance of a cold winter than a warm one.”

Anyway, Delingpole gets someone to do the numbers for him, and finds that the odds of three severe winters is 1 in 8000 (presumably because that’s 1 in 20, to the power of three – not really the right way to do the sum in this case anyway), therefore the Met Office must be wrong. The old joke comes to mind of a guy who fires blindly into a wall, looks for a spot where several bullets have clustered together, draws a bullseye around them and declares himself a crackshot. Most winters of course are not severe, and two or three severe ones clustered together among mild ones, while a bit unlikely, isn’t as impossible as Delingpole claims, and itself shows nothing. If lots of winters were severe over decades, then that would prove the 1 in 20 winters is severe claim was inaccurate. On their own, three winters prove nothing. If you throw a coin and score three heads in a row, that doesn’t prove the coin is loaded, especially if the last few tosses had a good mixture of heads and tails. If you score a hundred heads in a row, then you should be suspicious.

Okay, that was a diversion, since the Met Office’s predictions had little to do with global warming. Anyway, remember those eco-fascists Delingpole was complaining about?

Heads are going to roll for this, they’ll have to. But however many heads do roll it won’t be enough. Always remember this: the Warmist faith so fervently held and promulgated by the Met Office is exactly the same faith so passionately, unswervingly followed by David Cameron, Chris Huhne, Greg Barker, the Coalition’s energy spokesman in the Lords Lord Marland, and all but five members of the last parliament. And also by the BBC, the Prince of Wales, almost every national newspaper, the European Union, the Royal Society, the New York Times, CNBC, the Obama administration, the Australian and New Zealand governments, your children’s schools, our major universities, our minor universities, the University of East Anglia, your local council….

Truly there just aren’t enough bullets!

Giving an airport a certificate? Fascist.

Shooting politicians, journalists, teachers and scientists because you disagree with them? Not fascist.

Good to know.

,

6 Comments

The Mail says is the IPCC is wrong. In fact…

Today’s misleading climate change story, courtesy of the Mail: Alarmist Doomsday warning of rising seas ‘was wrong’, says Met Office study.

Alarming predictions that global warming could cause sea levels to rise 6ft in the next century are wrong, it has emerged.

The forecast made by the influential 2007 Intergovernmental Panel on Climate Change, which would have seen cities around the world submerged by water, now looks ‘unlikely’.

A Met Office study also rules out the shutdown of the Atlantic Ocean’s conveyor belt, which would trigger Arctic winters in Britain like those seen in the film The Day After Tomorrow.

Okay, before I point out everything else that’s wrong with this: The Day After Tomorrow is about as accurate a depiction of climate change as The Matrix is of computer programming. Regardless of what the climate does, we won’t actually have entire cities buried under frozen tsunamis, roamed by packs of wolves.

So, was the IPCC wrong? In their AR4 report (the one in question here) they said that over the 21st century, sea levels will rise by between 18 cm and 58 cm. In Daily Mail terms, that’s between 7 inches and about 2 feet (p. 45, which is p. 23 of the pdf, annoyingly). A 2 foot rise this century is rather a lot less than a 6 foot rise this century. Where did this error come from?

According to the report, beyond 2100, sea levels could rise higher – the theoretical maximum looks to be something between 7 and 10 metres – but that’s not what the Met Office study, “Summary of Post-IPCC AR4 work“, is about.

The Met Office says:

  • The relationship between temperature and sea-level rise is non-linear and the range for 21st century sea-level rise remains uncertain.
  • Some evidence that sea-level rise by 2100 may exceed the 95th percentile AR4 model-based projection of 59 cm.
  • Evidence that a rise significantly above 2 m by 2100 is very unlikely.

Median projections for 2100 under ‘business as usual’ scenario: AR4 model range of sea-level rise for this scenario was 0.21–0.59 cm. However, some of the newer evidence suggests that a sea-level rise of 2 m cannot be ruled out, but an increase of more than 1 m is currently viewed as unlikely.

So in fact, the Met Office says that the IPCC may have underestimated sea level rises, not overestimated them.

The Met Office never called the IPCC “alarmist”, they never said the IPCC “was wrong”, and even if the upper limit of an IPCC estimate had been rounded down, that wouldn’t make it wrong. It would just mean that we were to able to put more accurate upper bounds on sea level rise. This is all quite clear in the Met Office study; surely “Daily Mail Reporter” has to have read this report – and the IPCC report – to write this story. There’s no way such a massive error could have crept into this story.

The Mail accuses the IPCC of being alarmist and wrong. Perhaps they should look in the mirror first.

,

Leave a comment

%d bloggers like this: