Tuesday, May 30, 2006

Queue Rating

In today's column, Polly Toynbee writes:

Chris Smith is one of the few politicians to retire knowing he has done something brilliant - restoring free entry to museums and galleries, swelling attendances by 50%.
I hadn't appreciated that a seat in the House of Lords counted as retirement, but anyway. Swelling museum attendance by 50% looks like quite a good thing, until you consult the official figures. According to the ONS figures cited here, visits to museums and art galleries increased by 6% between 1997 and 2002. Social Trends 36 (the 2006 edition) says that between 2002 and 2003 they increased by 1% and between 2003 and 2004 they increased by 4%.

When we look at the Department for Culture, Media and Sport's website, we can find a press release lauding the increase in visitor numbers here. The figures revealed are fascinating -- particularly in Table 3 which is linked to at the end of the press release.

Charges affected fewer than 30% of museum visits back when the charges were scrapped in 2001. Since then, there has been a 67% increase in visits to museums which used to charge (69% in the first year, so actually a fall after the first year boost). Museums which were always free saw a 2% increase. Overall visits increased 21%.

And not 50%.

UPDATE: Also in the column, Polly writes:
Creative Partnerships was set up in just 36 deprived areas to bring artists of all kinds to work in 1,100 of the poorest schools.
No. Creative Partnerships was set up in just 16 areas. It was then expanded to 34 and subsequently 36. For detail, see the Creative Partnerships website or the, er, Guardian.

Friday, May 26, 2006

OFFS

In today's column, Polly wirtes about the by-elections to the Wyre borough Council in Lancashire. She writes:

Last Thursday, two weeks after the main local elections, there was a byelection in Park Ward, the poorest in Lancashire and the second-safest Labour seat in the county; activists piled in to support a good local candidate. But they were shocked to lose, with a staggering 27% swing to the Tories in a seat that was never Tory before.
Well, actually, the last time the ward was contested, in May 2003, the ward had two seats, so actually no-one had held that one single seat before (see, for example, here). Polly also writes that:
Labour is already delivering more babies: the ONS suggests a rise in the birth rate is partly due to mothers getting more help.
A nice pun, but an unsourced claim. I cannot find any reference to the ONS saying why the birth rate has increased -- indeed, their 18 May press release of the figures is just a bland statement of the facts. On the other hand, according to the, er, Guardian on 19 May:
"We looked at the reasons for this slight, hopeful rise," said Julia Margo, author of a report called Population Politics published in February by the Institute for Public Policy Research. "It seems to map on to 2001 when Labour started pushing on family friendly policies and childcare. There is a better deal now from government than ever before." But, she added, there were other possible reasons that do not hold out promise for the rise across all groups which is needed for a sustained increase in births. "We don't have access to the background data, which would tell us whether there are socio-economic differences, whether professional women will still be having less children." It could be that the rise is restricted to poorer women and those from migrant groups, who traditionally have had larger families.

So, it's not the ONS, which is a non-partisan government department, but rather the Institute for Public Policy Research which isn't. And it "seems to map" and they "don't have access to the background data." But otherwise an accurate summary.

UPDATE: In the column, Polly also writes that:
Surveying their staff at the Treasury, the officer class have just discovered that 88% of them have never worked for anyone but Gordon Brown.
This doesn't really seem to square with HM Treasury's Departmental Report of June 2005 (pdf link here) which says on page 47 that:
The Treasury encourages staff to gain outisde experience through secondments to both other government departments and the private sector. ... In 2004, 52 per cent of all new entrants were loaned or seconded in, and 16 per cent of the total staff in post at the end of 2004 were either on loan or seconded. Similarly 12 per cent of Treasury staff were on loan or seconded out to other organisations. [emphasis added]

Thursday, May 25, 2006

Better or Worse? p.190

More opening of the book to a random page.  On page 190 of Better or Worse?, Toynbee and Walker write:

"...by 2005 it [i.e. DFID] was pushing £9 out of every £10 in UK aid to low-income countries.  Half of DFID's bilateral aid went to Africa..."

No.  According to the DFID's Statistics on International Development 2005 (aka SID), in 2004/5 DFID's total bilateral programme was £2.14bn, of which £0.87bn went to Africa (source: Table 11 of SID.  pdf link here).  This is 40%, not half.

Similarly, £1.42bn went to low-income countries (see p.4 of the pdf linked above).  Of DFID's total bilateral programme, I make this about £6.59 out of every £10, and not £9.

But this is DFID's bilateral aid, whereas Polly's "£9 out of every £10" figure ostensibly refers to all aid, and DFID's bilateral programme is only a portion of all UK aid.  In 2003/4, for example, total gross public expenditure on bilateral aid was some £2.61bn (of which some £1.97bn was spent through DFID -- comparable to the £2.14bn figure for 2004/5 quoted above), whereas total gross public expenditure on aid of all kinds was £4.74bn (source: Annual Abstract of Statistics 2005, p.9.  The two series I am referring to are LUQL and LUPI).  The difference is made up of moneys given to multilateral agencies to spend, of which by far and away the largest proportion goes to the European Community for them to disburse.  Could they be spending so much money on supporting low-income countries to make Polly's figures accurate?  Sadly, probably not.  Says who?  Says Polly:

Gordon Brown was critical of the EU's aid programme which dispensed £6bn a year.  Instead of hitting the most extreme poverty, money flowed to favoured areas, notably the Balkans...

This quote is from Better or Worse?, p.190.

Better or Worse? p.78

In a quiet fifteen minutes, I have discovered a new game -- opening my copy of Better or Worse?, the 2005 book written by Polly Toynbee and David Walker, to a random page and seeing how many dubious facts I could find.  This morning, it was page 78 (my copy is a Bloomsbury paperback).

I found this quote:

In the media they [apparently newspaper editors and BBC directors] controlled, they reflected a world of privilege -- so a threat to private school education or private health was wrongly described as an attack on 'middle England'.

When I Google "attack on middle England", I get seven results - maybe the problem is not as severe as Polly thinks?  Of the seven, four are from the media (two are from Hansard, and one from Amazon).  Of the four media citations:

  • One was from the BBC in 2002, talking about the tax increases in the 2002 budget.  Note that the 2002 increases were in National Insurance contributions, and were not threats to private education or private health per se
  • One was from the, er, Guardian, and quoted Gordon Brown denying that the national insurance increases in the 2002 budget were an attack on middle England.  No reference to private health or education
  • One was from the Mirror, and which contained this curiously ungrammatical sentence: "Who view every tax and welfare payment as a financial attack on Middle England." [emphasis added]  If you go and read the piece in the Mirror (it's OK, you don't have to) it seems to be referring to John Redwood.  Note, again, no reference to private education or health
  • The final quote I can find from the UK media is this, from Money Marketing Live: 'One of the big Budget surprises was Gordon Brown's attack on Middle England with his proposed inheritance tax changes on trusts. These are considered by many to be ill-thought-out and totally unjustified. They are retroactive, to use Government terminology, and can apply to existing trusts in certain circumstances.' John Wolley, MM, 20 April 2006."  Given the quote is dated after Polly's book was published, she can't really be referring to this.  And it doesn't refer to private health or education

So, four quotes from the media, none of which refers to private education or private health.

Incidentally, for those interested in Polly Toynbee's salary, consider the sentence which comes immediately before the one I quoted above, which I reproduce in its entirety here:

Newspaper editors and BBC directors and everyone they knew in their hermetic worlds all earned many multiples of £100,000.

I take this to mean that Polly must be earning many multiples of £100,000.

My other big problem with page 78 comes with these two sentences:

Professor Richard Layard showed how people draw a sense of their esteem and worth from their relative place in a nation's pecking order. Asked a hypothetical question, most say they would choose to have less money in a society where everyone had a fairer share, than more absolute wealth in a society where everyone else was far richer.

As well as getting her facts wrong, another Polly Toynbee trait is to quote apparent facts without a source, making it difficult to judge whether the fact is accurate or not -- as happens with the "hypothetical question" she refers to above.  So, with no real guidance as to where it came from, I am assuming it comes from Layard's work, a lot of which is posted here(*).  The closest thing in Layard's work which I can find to Polly's hypothetical question is this little experiment:

A sample of Harvard graduate students were asked:
1. Which of these two worlds would you prefer? (Prices are the same in each)
    A. You get $50k and others get half that.
    B. You get $100k but others get more than double that.
2. Which of these two worlds would you prefer?
    C. You get 2 weeks holiday and others get half that.
    D. You get 4 weeks holiday but others get twice that.
The majority answered A to question 1, and D to question 2

This is taken from "Towards a happier society" (pdf link here).  I may be wrong, and this is not what Polly is referring to -- and I will withdraw this critique if this turns out to be the case (though not my criticism about inadequate sourcing).

My first problem is that Harvard graduate students, lovely though they may be, are hardly a representative sample of the UK population.  Secondly, the two scenarios described in A and B are hardly different in their levels of inequality -- in both cases they seem pretty equal societies with one outlier, the respondent.  This is not a test of whether we want everyone to have "a fairer share" or not.

* At the time of writing, the links on this page to Layard's work are broken, as they have a spurious "www" in the link.  You'll have to click on the link, and then delete the "www." from the address in the address bar of your browser.

Tuesday, May 23, 2006

Apologies again for the late discovery of today's column by Polly; I could not find it through www.guardian.co.uk's search feautre, and actually still can't as at the time of writing. It does, however, appear on CommentisFree here. She starts by saying:

Here global warming is measured by how often the steel gates are closed; in 1987, it was only once every two years: now it's four times a year, eight times more often.

You can see the detail here -- the figures are a lot noisier than you might think from this description. In fact, the barrier had to close six times in 1990, and not at all in 1991. Nine times in 1993 and only once in 1994. Only twice in 2004 and 2005, though this comes after 18 times in 2003.

Later on she says:
If in 1987 the prudent designers of the Thames barrier built in expectation of global warming...
Actually, if the designers of the Thames Barrier were building anything in 1987, I hope it was a time machine. According to the Environment Agency, "it becomae operational in 1982" (source). The Thames Barrier, I mean, not the time machine.

Of Thames Water, Polly says:

targets for fixing leaks have all been missed.
Strangely, when we look at Ofwat's "Security of supply, leakage and the efficient use of water 2004/5" report (pdf link here), and the section on Thames Water (pages 39-40), we find quotes like:
The other positive aspect of the company’s performance in 2004-05 was that its area outside London had leakage performance in line with targets and at a level comparable to other companies in England and Wales.
and
Thames’ quarterly progress reports actually showed it to be on target until a late winter leakage spike at the end of February 2005...
and
This was the first objective achieved.
I am not a big fan of Thames Water and their performance in finding and fixing leaks, but the picture is a little more nuanced than we are led to believe.

Not knowing your enemy

No column in the Guardian today, but I notice that Polly has contributed to the May/June issue of Progress magazine in the form of a column called Any Questions. From that column come this question and answer:

If you were able to spend an hour with one dead, historical figure, who would it be and what would you ask them?

Abraham – Could you be dissuaded from founding three world religions which will cause more bloodshed and despair for more millennia than you could possibly imagine?

Founding? Really? Inspiring, maybe. Being a central figure in the development of, maybe. But founding? Is it really believable that a day or two after his death one could have asked any of his contemporaries what they thought of Christianity or of Islam?

UPDATE: There now appears to be a Polly column on CommentisFree.com, though not on the main Guardian site. Apologies, fact check to follow.

Friday, May 19, 2006

Pollyannomymous

In a post today on CommentisFree, Polly writes of people who criticise her anonymously:

Tell me something else, how many of you bother to buy the Guardian? Here we are, the only non-profit paper with no megalomaniac owner, like all newspapers in need of paying readers at a time when the press is in decline.
Given that the paper loses money and the website makes money, I would be positively encouraging the bloggers and discouraging people from buying the paper. But then I am a sucker for what Polly calls "Adam Smith's hidden hand of profit" and the rest of us call the invisible hand.

Incidentally, I think the broad swipe of today's article, lumping together people who quibble with Polly's gift for inaccurate precis with people who bombard her with ad hominem abuse quite an underhand attempt to deflect criticism.

Unclear

On days like today, I feel disappointed I elected to critique factual accuracy rather than, say, writing style. In today's column, potatoes are launched, and herrings hijack. She manages to speak to the energy minister, Malcolm Wicks, and she writes:

So I asked Malcolm Wicks the vital question.
The vital question, it turns out, is this:
Will nuclear power get any special inducement not offered to other forms of energy generation? Will there be a genuinely level playing field giving every prospective form of clean energy the same chance to prove its viability? That means nuclear power stations would have to pay not only for their waste storage, but the high cost of full insurance: currently they only cover themselves up to a paltry £140m of risk - so a Chernobyl would leave the state picking up a huge bill for compensation and clean-up. Will future nuclear generators be forced to pay into a fund each year enough money to cover all their own decommissioning? The state is now paying a £70bn bill to close existing stations - with the price still rising.
127 words, five sentences and three questions. Bafflingly, Wicks replies "yes" four times to three questions. I also enjoyed the setup to the interview with Wicks, which was:
Bumping into him yesterday, he gave a wry shrug

The image of someone shrugging wryly as they bump into themselves will stay with me for a while.

I am also disappointed that I choose to quibble on facts rather than on logic, as the triumphant glee with which Polly suggests that people won't invest in nuclear power stations because the benefits are public policy ones rather than private ones can surely not be the same Polly Toynbee who wrote in her previous column that
money doesn't buy as much happiness as common social goods. Taxes are good value: health, education, the arts, parks or sports are more precious and pleasurable than anything bought in a shop.
Incidentally, I value food -- which, tediously, I sometimes buy in a shop -- above sports.

But no, I had to go and set myself up as someone who critiques the (non-)facts. Luckily this is not onerous.

She writes
Writing yesterday in the Financial Times, Alexander Johnston, from a leading consultancy to Fortune 100 companies, lays out the terms he thinks investors would require.

I presume she is referring by "leading consultancy" to Arthur D Little, on whose advisory board Johnston sits. Arthur D Little is (no offence) not a leading consultancy to Fortune 100 companies. It has had a fair share of economic trouble, having filed for Chapter 11 bankruptcy protection in the US and having to sell itself off (see, for example, this news story). Its website currently shows only two US office locations -- Boston and Houston. I am sure it is a fine company full of worthy professionals, but it is not, say, a McKinsey.

Polly also writes:
Look how simply labelling white goods with energy-saving ratings made virtually all of them AAA in a short time.
I think she may mean new white goods sold (unless she thinks the labelling scheme has somehow magically made washing machines bought in the 1980's more energy efficient). Even given this generous leeway, it's not strictly speaking true. As a quick check, I had a quick look at the washing machines John Lewis offers on its website. The first four on offer were:
  • John Lewis JLWM1202 Washing Machine, White
  • AEG L16830 Washer Dryer, White
  • Miele WT2670 Washer Dryer, White
  • Zanussi ZWF1421 Washing Machine, White
The energy ratings, respectively, were: A+; B; A+; and A.

I know this is not a rigorous empirical test, nor is it statistically robust. But it is enough for me to not believe that "virtually all of them [are] AAA".

Tuesday, May 16, 2006

Ironic quote marks?

Today's column is relatively free from fiction masquerading as facts (they are, of course, all relatively free of facts). However, Polly makes up for it by mangling a quote. Consider this quote from Pollyanna's column:

Professor Roger Silverstone, of the LSE, calls the research hopeful, "revealing the emergence of a dynamic, socially engaged and environmentally conscious" voter - not a few, but 20 million.
Now compare that with (and most certainly not to) the actual Silverstone quote from this BBC page:

Roger Silverstone, Professor of Media and Communications at the London School of economics [sic] says: "This is, in many respects, a significant and indeed a hopeful piece of research.

"It reveals the emergence of a dynamic, socially-engaged and environmentally conscious consumer at the heart of British culture which should have real consequences for the ways in which commodities are bought and sold, and media are consumed."
We can skip lightly over Polly's "revealing" and the original "reveals" -- this is the sort of attention to detail that we only expect from credible journalists. However, the substitution of "voter" for "consumer" is quite a big one, particularly as Silverstone goes on to talk about the consumption of commodities and media, and not about voting intentions.

I understand, of course, why Pollyanna does not quote one of the findings from the research -- that "86% of Big Britons claim they find it increasingly difficult to trust what they read" (you can find the quote on the BBC website linked to above) -- given her own manipulation and fabrications.

Friday, May 12, 2006

Right to lie?

Today's column, on the right to die, contains this quote:

In the polls, over 80% support the right to die and have done for the last 25 years.
The best poll I know of in the UK for measuring changes in social attitudes is the British Social Attitudes Survey. Care to guess what they have to say about the issue? Well, they report that:
Opinion polls show that not only do euthanasia and assisted suicide already enjoy the support of a substantial majority of the UK population, but also that this support is actually growing. A 75% majority in favour of permitting medical assistance in the ending the life of a sufferer from a painful, incurable disease in 1984 increased to 79% in 1989, and 82% in 1994

This is from British Social Attitudes Report, 1996. Now I appreciate not every library has a copy of the tome. It is, however, quoted here.

Even Dignity in Dying, referenced by Pollyanna in the column, have on their website the historical results of polls, showing that support did not exceed 80% until 1994.

Needless to say, the facts do not bear out Polly's contention that an 80% majority have supported the right to die for over 25 years.

Tuesday, May 09, 2006

Criminal statistics

Polly claims today, in a piece about the recent local elections, that we should:

Take Hammersmith and Fulham: it boasted the biggest fall in crime
It is a claim she has made before, and it wasn't sourced or qualified then either.

At the time of writing, as was the case before the local elections, this claim is not supported by the Metropolitan Police's Crime Figures website which, as of 9 May 2006, shows that Hammersmith & Fulham had seen a 4.7% decrease in crime on a rolling 12-month basis.

Good. Very good, even. But not as good as Barnet's 8% drop, Sutton's 5.1% drop, Wandsworth's 4.8% drop, or Lambeth's 8.3% drop, and so on.

Of course, it is probably possible to find or construct some measure of crime on which Hammersmith & Fulham has seen the largest fall over a carefully chosen period. However, due to Toynbee's cavalier use of the facts we don't know whether she has carefully chosen the definition and the time period to suit her thesis, or whether she has just made the statistic up.

Sources, please!

Friday, May 05, 2006

Copying from the Telegraph

Polly Toynbee's column today is relatively fact-free (well, they always are, but at least this time there aren't a lot of non-facts masquerading as facts). however, she does say:

Indeed, recent research on child care found that children left with grandparents all day did worse than children in good nurseries.
As ever with Pollyanna, there is no source, but at a guess she is referring to an article called "The Effects of a Mother's Return to Work Decision on Child Development in the UK" by Paul Gregg, Elizabeth Washbrook, Carol Propper and Simon Burgess, published in the Economic Journal of February 2005 (pdf of the article is available here at the time of writing). The article does not distinguish between care provided by "grandparents" and "good nurseries" -- it distinguishes among: informal unpaid care by a relative; non-relative paid care; and centre-based care. The research finds that relying on unpaid care by a relative (as opposed to paid care) is detrimental to children where the mother returns to work when her child is 18 months old or younger, though only, bizarrely, in better off households. Why? Difficult to say. Quote from the article:
We conclude that the use of predominantly relative care is damaging only for children in the more advantaged households and that children in less advantaged households are not harmed by early full time maternal employment. However, we are unable to say much about the reason underlying this result.
This is so far from Toynbee's inaccurate précis that you would certainly be entitled to ask why I think that this is the research to which she is alluding. Fair question. It rather seems as if the mangling starts, strangely enough, in the Telegraph. In this column, their social affairs correspondent Sarah Womack summarises the research thus:

Mothers who return to work part-time when their children are as young as three months old have no adverse effect on the future development of their offspring, according to the latest academic survey.

But full-time working mothers who leave their children under 18 months old in the sole care of grandparents or of friends risk seeing their children fall behind at school.

Researchers say there are "significant" detrimental effects when the child is left with an unpaid carer.

The child is, on average, three months behind his or her peers by the age of seven, doing less well in literacy and numeracy tests and being less adept with language.

If the child goes to a nursery rather than grandma's while his or her mother works full-time, he or she is about one month to two months behind by the age of seven.

Note that it starts as an accurate synopsis -- the second paragraph is not bad (though "of grandparents or of friends" is not the same as "relatives" -- the authors of the article talk about, for example, partners providing care). However, by the final paragraph of the quote Womack is trying to give examples and straying dangerously from the research she is trying to summarise. However, she is not straying as badly as Toynbee's sloppy, innacurate one-sentence summary.