Archive for category Damned lies and statistics
As you may have heard, there’s currently a climate conference taking place in Cancún, Mexico. As you may also have heard, it’s snowing.
If, last week, frozen behind a snowdrift, you heard a faint hysterical squeaking, it might well have been the sound of those 20,000 delegates holed up behind a wall of armed security guards in the sun-drenched Mexican holiday resort of Cancun, telling each other that the world is more threatened by runaway global warming than ever.
Yes, it’s the classic “If global warming is real why is it cold?” argument from Christopher Booker, blithely ignoring the fact that 2010 is set to be one of the hottest years on record (and 2000-2009 was the hottest decade on record) regardless of what the weather is like for one week on a small northerly island. To be fair, this isn’t his only argument (even if he does return to the subject three more times in the space of the column). After all, he also has “If the oceans are getting more acidic, why are they still alkaline?”
Far from the oceans acidifying, their pH currently ranges between 7.9 and 8.3, putting them very firmly on the alkaline side of the threshold, at 7.0.
Of course, the fact that the seas are alkaline doesn’t mean they can’t be less alkaline than they were years ago. And sure enough, according to the Australian Antarctic Division:
CO2 from human activities has caused the pH of ocean surface waters to drop by 0.11 pH units. This might not sound like much, but it is equivalent to a 30% increase in acidity.*
Ocean acidification is never going to turn the oceans completely acidic – that would require a 1580% rise in ocean acidity, which is a bit unlikely. What acidification will do – and what it has already done, in fact – is bring acidity up just enough to interfere with creatures like coral, which rely on the precise chemistry of the ocean to produce their skeletons.
His other argument?
It is only those same old computer models that predict that Tuvalu and the Maldives are about to drown, when real measurements show the sea around them not to be rising at all.
It’s true that one dataset from Tuvalu did appear to show no sea level rise. Unfortunately (for both Booker and the Tuvaluans) that data was taken from a single station over a relatively short period and presented without uncertainties, making it effectively meaningless. Analysis of more data from Tuvalu (PDF, p.11) finds a sea level rise of 1.2 ± 0.8 mm/year. The uncertainty in the data is still quite large (not least because the island of Tuvalu itself is moving by a small but unknown amount), but there certainly appears to be a sea level rise. At any rate, regardless of whether or not the sea level is rising at Tuvalu right now, it’s certainly rising worldwide at a rate which threatens people living in low-lying land.
The global warming scare may have been fun for the children while it lasted. But the time has come for the joke to be declared well and truly over.
Incidentally, Booker finishes off his column with a link to some people singing Handel’s Hallelujah in a food court which, he claims, is “the very opposite of all that is stood for by global warming, social workers, the European Union, the Coalition Government and the rest of this column’s usual fare“. Because if there’s one thing environmentalists really hate, it’s people singing.
* pH is a logarithmic measure of acidity, which means that numbers appear to work slightly oddly. A drop of about 0.3 on the pH scale corresponds roughly to a doubling of the activity of H+ ions (the ions that cause acidity), and a drop of 1 (from 8 to 7, for instance) represents a tenfold increase in H+ activity.
The idea that the Daily Express would somehow be able to change Britain’s timezone was stupid. The idea that the Daily Express can single-handedly tear Britain out of the EU however is just plain cute. It’s a bit like watching a puppy try to take part in an Olympic marathon. You know it hasn’t a hope in hell of succeeding, but simply the idea that it thinks it can win makes you go “Aww” against your better judgement.
Just as last time, they claimed that 29 million people supported them based on a survey of a few thousand, the Express is once again overstating support for its crusade – which I remind you only started today – by quoting a few Eurosceptics and claiming that they represent “a huge groundswell of support … gathering behind the Daily Express Crusade“, and that the Daily Express running a front page piece about how it doesn’t like Europe is “a turning point in the battle to win back Britain’s independence“.
Edit: Now they’re claiming that “99 per cent of people agree we should quit the European Union“, when of course what they actually mean is that 99% of Express readers with a strong enough opinion to ring a premium rate number buried somewhere in the newspaper agree. Via Primly Stable and Enemies of Reason.
Anyway, as you might expect, the Express gets maybe a wee bit overenthusiastic, claiming that unemployment is high in EU because politicians are for some reason deliberately sabotaging recovery with “new job-destroying regulations” and running with a “what have the Romans ever done for us?” argument that “Almost nothing the EU has proposed or enacted has benefited Britain“.
Perhaps the ‘best’ part of the article though is when they move past mangled memories of Black Wednesday and the claims that the only difference between us and Switzerland is EU membership, and move onto history. Now, the Daily Express is always complaining that schools aren’t teaching history properly, but instead twist it to their own ideological ends. Well, I’m sure the Express won’t stoop to that lev…
The creation of the EU is explained by the perfectly understandable desire to avoid further conflict on a continent that had been the scene of two world wars.
But Britain is a land apart: A precious stone set in the silver sea, as Shakespeare so evocatively put it; a realm with a glorious island story stretching back a thousand years, with links to every continent and a language taken up throughout the world.
You know, completely unlike a France, a realm with a glorious story stretching back a thousand years with links to every continent and a language taken up throughout the world.
Anyway, I’m surprised the Express, usually rather big on the whole remembrance thing, forgets which country suffered massive economic damage as a result of those two world wars, and therefore may have a bit of an interest in preventing another one happening. Here’s a hint, it was us.
This is mostly just a a collection of thoughts that were too long for Twitter.
In an interview with Attitude magazine, Stephen Fry claimed that women do not enjoy sex, saying “The only reason women will have sex with [men] is that sex is the price they are willing to pay for a relationship with a man, which is what they want“. The evidence he gave to back up his assertion?
“Of course a lot of women will deny this and say, ‘Oh, no, but I love sex, I love it!’ But do they go around having it the way that gay men do?
“Gay men are the perfect acid test. If they want to get their rocks off, they go into a park where they know they can do it.”
Well, luckily, there’s a nice big data set we can use to test Fry’s claims. Recently, the dating site OKCupid performed a survey of its users to analyse the how the dating habits and sex lives of gay and bisexual users varied from those of straight users. Obviously this will not be a perfectly unbiased survey – people who are members of dating sites are perhaps more likely to be looking for sexual partners than average, for instance. Still, it’s a very big sample – 3.2 million people in fact – and the bias should affect gay people as well as straight people.
If Stephen Fry is right, we should find two things. Firstly, that gay men would have way more sexual partners than straight women (and indeed straight men), and secondly that gay women would essentially be celibate. After all, if women don’t enjoy it, lesbian couples have no reason to have sex.
In fact, here are the results. Gay men have had, on average, 6 partners. Straight men have had, on average, 6 partners. Gay women have had, on average, 6 partners. Straight women have had, on average, 6 partners.
There is no statistically significant difference between the sex lives of men and women, nor between gay people and straight people. Stephen Fry’s comments are simply not backed up by the science.
Edit: Stephen Fry claims to have been misquoted. The journalist who interviewed Fry on the other hand has said “he delivered [the comment] with certainty and it was clearly something he’d thought about.” It’ll be interesting to see who turns out to be right.
SUDDENLY, the nights are drawing in. The sky is filling with the smoky grey clots of churning vapour that herald the annual return of Boreas and his frigid kingdom of shade and bluster. Worst of all, men, women and children are forced to eat their tea in the dark because no-one has invented the light bulb yet.
From the darkness rises our saviour, St. Desmond, at the vanguard of the heroic Crusade for Change. Where the Daily Express leads, an army 29 million strong follows…
Wait, 29 million? That sounds a bit much, surely? The turnout at the last general election was only 29.6 million – are you telling me as many people care about the Daily Express‘s “crusade” as care about national politics in general?
In fact, all that happened is that Santander carried out a survey, and 58% of those polled said that the government should look at the current Summer Time system. Not that they definitely agreed with the Express, just that they thought it might be worth checking. The Daily Express has then multiplied this figure by 50 million – a ballpark estimate of the population of England – and assumed that therefore 29 million people must be in favour.
Without being able to see the survey, there’s no way of knowing how representative of the British population this survey is. Already though, one thing seems clear; while the Express talks about 29 million Britons, it might be more accurate to say 29 million Englanders. Messing with British Summer Time is somewhat less popular in Scotland and the North of England, as the Daily Express itself secretly recognises.
Let’s look at some of the other findings of the study.
According to the study, 45 per cent will feel unusually depressed during the daylight-starved winter months.
Concerns are also mounting about children walking home in the dark and the danger of personal injury, with one in four people saying they feel more at risk as evenings draw in.
In addition, some 36 per cent – 17.7 million people – believe there is an increased chance of road traffic accidents, and one in four also insist they feel more at risk from burglary.
In other words, 3 in 4 do not feel more at risk, 64% of people did not say they believed there was an increased chance of road traffic accidents, and 3 in 4 do not feel at risk of burglary. Bringing seasonal affective disorder (winter depression) into this is a low blow – it appears to be caused not by clock changes, but just by the fact that there is less light overall during winter.
I’m not sure how much difference clock changes would have to some of those anyway. There is good evidence that “double summer time” reduces traffic accidents in England (but raises them in Scotland), but I couldn’t find any evidence about the others – there are some papers on whether clock changes can cause depression, but the conclusions look, well, inconclusive.
So, in the interests of improving the state of mathematics, I’m going to poll the entire world about whether the Daily Express needs remedial lessons on how to use statistics and surveys. That means potentially upwards of 6,877,939,067 voters, all weighing in. Lend us your opinion, have your say, and help my Crusade for Change!
(Hat tip to Exclarotive)
The papers aren’t even trying any more. Once upon a time, they might at least pretend they weren’t simply churning out TaxPayers’ Alliance press releases; today’s articles “Council EU jobs cost £41m” (Express) and “Councils spend £41million a year on non-jobs” (Telegraph) however are utterly shameless about it.
First things first: the TaxPayers’ Alliance report (eurgh, I guess I should link to it) found the cost of all “unnecessary” jobs – in other words, any job Richard Littlejohn might disapprove of – was £41 million. That’s not just European Officers, but Diversity Officers, Political Advisers and Climate Change Officers* too.
The report contradicts itself several times: the TPA point to the fact that different councils employ different numbers of staff as an indication of waste – “the disparities across councils are evidence that the hiring of specific staff is not necessary” – only to later state “As proponents of localism, the TaxPayers’ Alliance believes that local authorities should be granted as much freedom as possible to make decisions that benefit local taxpayers. This will inevitably mean that councils will pursue different policy objectives“. So councils should have as much freedom as they like as long as they only use that freedom to do exactly what the TPA wants?
That damn war on motorists, eh? Not only did they put a bus lane on the M4 (which actually improved traffic flow significantly), now they’re putting 20 mph speed limits on roads! Naturally, the Daily Mail is unhappy, and now they have the ammunition to take these speed limits down! M… maybe.
Why death rates INCREASED in 20mph zones… and getting rid of cameras reduces accidents says today’s Mail. Now, you might think that with the headline beginning with the word “why”, that mean the Daily Mail was going to tell us why death rates increased. Of course, they don’t. That might be because death rates didn’t really increase at all in any meaningful sense.
(Note: Daily Mail links now go via the wonderful istyosty service, which means that the Mail does not get the advertising revenue from the hits)
The word “uncertainty” has a special meaning in science, quite different to its normal everyday meaning. If I say that I am uncertain about my future, or what to have for tea tonight,* then it might mean that I am “undecided” or “unsure”. If however I do an experiment and I say that there are uncertainties in my data, that does not mean I am undecided or unsure about the results, or that the whole thing was a waste of time.
Instead, all it means is that I realise that my measurements are not – and can never be – 100% precise and accurate. For example, my kitchen scales weigh things in units of grams, so they could tell me the difference between 99 grams of flour and 100 grams, but aren’t precise enough to tell me the difference between 100.1 grams and 100.2 grams. That’s an example of an uncertainty (±0.5 grams in this case), but it’s not one that means that the scales are completely useless; they might not be perfectly accurate but I can still use them to bake a cake.
I say this because apparently the Daily Mail‘s science editor apparently fails to grasp this fairly simple idea:
James Delingpole, the only mainstream journalist to blame global warming on the Bilderberg Illuminati New World Order, has an urgent message: the foundations of Britain’s wind farms are about to collapse!
Where did he get this horrifying news? Well, an anonymous source emailed him a rumour heard from an another anonymous source. Apparently, it’s all extremely serious, which is why the mainstream media, which as we all know absolutely loves wind turbines and will defend them at any cost, is covering it up.* Still, as his source says, that’s what happens “when an industry builds bigger and bigger things without having built up a history of experience on the ones built earlier” since of course the “foundation” is a radical new technology and no-one ever built a tall steel structure before wind turbines came along.
Because even Delingpole admits a single unsourced email isn’t enough to base an entire column on, he also quotes, unchecked, Christopher Brooker’s assertion that nuclear power gives you 13 times as much as power per pound as offshore wind – in fact, it’s only about 1.5 times as much, according to yesterday’s linked Telegraph article, and nuclear power is actually 11% more expensive than onshore wind (and of course, the cost of offshore wind is expected to fall). Apparently his source is a trustworthy-sounding blog called “EU Referendum” (I’m not going to link it – if you want to read it, it’s linked from Delingpole’s article), which in fact claims that for the £1.2 billion that it will cost to run the new 300 MW Thanet windfarm for the next 20 years, you could build a 1 GW nuclear power plant (in itself a dubious claim – exact figures are hard to come by, but this report (PDF, p.32) on the new Hinkley Point C reactor suggests a cost of £4 billion for a 1.6 GW reactor, or about £2.5 billion per gigawatt, twice what Brooker suggests).
Comparing the 20 year running cost of a plant with the construction cost of another is quite obviously misleading. In fact, Thanet cost £900 million to build, which works out at about £3 billion per gigawatt; slightly more than nuclear power, but not much – certainly not 13 times as much. Taking load factors – the percentage of maximum output that the power station usually produces – into account will admittedly bump this up – using the statistics from Renewable UK** it looks like this load factor adjustment brings the cost of the power roughly in line with the findings of the UK Energy Research Centre report from yesterday.
* Delingpole’s claim that “blinkered” mainstream journalists are refusing to criticise wind farms in fact puts me in mind of the Daily Mail worrying that “you can’t talk about immigration“… while devoting acres of newsprint to immigration. Yep, no-one’s talking about the cost of wind power.
** Incidentally, it’s worth reading the Renewable UK site for another reason too – it explains why it’s not really a concern when the wind stops blowing. In a nutshell, other power sources are themselves unreliable – even coal and nuclear plants can break down, after all – and so we need to keep power stations running on “spinning reserve” anyway. These reserve power stations can absorb the slack when the wind isn’t at full strength without producing significantly more pollution.
Edit: Just noticed the picture Delingpole uses of a fallen turbine, with the caption “One down…”. From that, you might think that these mythical foundations have already started to crack… except that in fact that turbine quite clearly fell because the tower bent in high winds, and its collapse had nothing to do with the foundations (and it’s the only turbine to have collapsed so far in England – “one down” indeed).
It’s a shame that the PCC refuse to rule on misleading headlines, because the Daily Mail has a doozy today: Revealed: Wind farm power twice as costly as gas or coal.
If you read that headline, you might assume that means that all wind farms cost twice as much to run as fossil fuel plants. Certainly, that’s the impression the Mail wants to leave you with but, as always, the devil is in the details.
The true cost of Britain’s massive expansion of wind farms has been revealed.
It costs nearly twice as much to generate electricity from an offshore wind farm as it does from a conventional power station, a scientific report has concluded.
And while the price of wind power is expected to fall in the coming decade, the researchers admit there is a slight chance it could rise even further.
Did you spot the giveaway word in there? Yep, in fact this information only deals with offshore wind farms – though the Daily Mail seems to forget this several times and refer to “wind power” unqualified . According to The Telegraph‘s slightly more accurate report, a megawatt hour of offshore wind power costs £149, a megawatt hour of onshore wind power costs £88 and a megawatt hour of coal power costs £80 (though I’d imagine that refers to old-fashioned coal power, not “clean” coal).
The “scientific report” in question is Great Expectations: The cost of offshore wind in UK waters by the UK Energy Research Centre which didn’t conclude that it costs twice as much at all. Instead, working from the initial data that offshore wind power was expensive, the researchers conclusions were that the costs of offshore wind power have been inflated by rising commodities prices, supply chain problems and the shift of the pound against the euro and that government action (encouraging turbine companies to manufacture parts in Britain, building larger ports capable of accepting the gigantic equipment) should be able to bring the prices down.
In other words, a more accurate headline might be “High costs of offshore wind set to fall”, though of course that would involve realising that this is a complicated issue, with lots of economic forces at work, instead of simply saying “Wind power is more expensive! And what do you do when the wind stops blowing, eh?”
Incidentally, the Daily Express‘s take on the report is even more bizarre:
THE £4billion of wind turbines off Britain’s coast still generate less electricity than a conventional power station, a report revealed yesterday.
And the cost of building offshore wind farms has doubled in the past five years to £3million per megawatt.
The UK’s offshore windfarms generate 1,341 MW. I can only find American data, sadly, but according to that, the average output of a coal power plant is 667 MW. So in fact, our offshore wind is generating as much as two average conventional power stations. I suppose by “a conventional power station” they could mean a monster like Drax or Didcot, but in that case the Express is being very misleading.
Most papers today cover news that the Integrated Household Survey found roughly 1.5% of the UK population self-described as gay or bisexual. There’s a nasty undercurrent to most of the articles, although only the Daily Express makes their point explicitly:
But critics have said it raises questions about the importance placed on homosexuality.
Tory MP Philip Davies said: “An awful lot of focus in diversity issues is given to people’s sexual preference and this difference is not quite as widespread as believed.
“That said, I do not see what relevance it is to anyone else. Someone’s sexual preference is a personal matter and it calls in to question why anyone is bothered at all.”
Yes indeed! If we can just fiddle the statistics to downplay the number of LGBT people, then we can just sweep them under the carpet and stop being bothered with such irrelevant things!
So before I go any further, I’d like to point out that even if this survey is correct, and there are just 700,000 gay people in Britain, not 3.5 million, that wouldn’t mean that gay rights would become less important or relevant, as The Express seems to claim.