Thursday, 11 July 2013

The Wisdom of Crowds - 47

The poor old Indie runs a piece this morning following research by the Royal Statistical Society and KCL that 'proves' that the public is 'wrong' on a whole range of social issues; for instance "Benefit fraud: the public think that £24 of every £100 of benefits is fraudulently claimed. Official estimates are that just 70 pence in every £100 is fraudulent - so the public conception is out by a factor of 34."

To a point, Lord Copper. It depends how you define 'fraudulently' - the researchers interpreting it in the strictest sense, whilst the public no doubt including 'undeservingly' in their definition. And the public's view that one-in-four on the dole needn't be is probably more accurate. Nil points, KCL. 

Likewise 'Teen pregnancy' - the public figure almost certainly includes single mums in their early twenties who may have been older than 19 when they gave birth but are included in the generic cohort . And immigration. And crime. 

In fact, all that Hetan Shah's little exercise proves is that on a sensible definition of social issues the crowd one again has the wisdom whilst the officious office-holder is exposed as a nitpicking disingenuist. No doubt there is research that 'proves' that only 0.87% of them actually are ....


JimS said...

BBC Radio 4's "More Or Less" programme included an item on 'Affordable Housing' some months ago. Apparently a house can cost £1M and still be 'affordable' if it is in the right area and been build under the right government scheme/scam.

The civil service delights in coining new meanings for common terms. It enables Permanent Secretaries and briefed ministers to be completely truthful before select committees whilst leaving their interrogators confused and perplexed as they are not familiar with the official 'code'.

My favourite one is 'value for money'. That just means that the department followed the contracting procedure which 'ensures' value for money is achieved (supposedly), whether or not it has is not the point.

Nick Drew said...

the prostitution of 'official statistics' is exceptionally annoying, because we need better than their damned lies

case in point - last week's hue-n-cry over health tourism: government says it doesn't know exactly what this costs the NHS but the best estimate is £30 million p.a.

£30m ? step into any inner-city A&E or maternity ward at any time. Sit in at any 'asylum-seeker' help-group and hear their stories. Look at the details of TB cases etc etc. Check out the number of unaccompanied minors that arrive on our shores and are cared for in every dimension at public expense. Whatever is the 'correct' figure, palming us off with "£30m" is an insult to our intelligence

no-one is seriously trying to quantify these things: when ministers ask, they get the brush-off from civil servants who have no intention of finding out

and so the "£30m" is used as the basis for why it's too small to matter

we eventually reach the point where some local authorities in the 'sheds-with-beds' belt resort to obtaining measurements of sewage output in order to get an estimate of the population in their area

a country that in the 21st C doesn't even have a reliable census, but does have a civil service determined not to improve matters, is in a rare state of anarchy

English Pensioner said...

Some years ago I was driving a car and became involved in a drink-related accident. No, I wasn't over the limit, but I hit a pedestrian who was. All motoring accidents where there is alcohol involved is recorded as "drink related", no doubt to produce a higher figure which is then assumed by the average person to relate to drunken drivers.

Demetrius said...

Nick Drew beat me to it. As so many "official statistics" or "estimates" are rubbish for a variety of good reasons(My stat's prof' was Allen)there is a good chance that the collection of personal experiences of those on the ground may well be nearer the mark.

Anthem said...

I suppose there is a difference between "fraudulently claimed" benefits and "necessarily claimed" or "legitimately claimed" benefits.

Whatever the stats say, when the welfare cost makes up the largest slice of the governmental spending pie, something is wrong.

It suggests that far too many people are incapable of living as people.

G. Tingey said...

Nick Drew
THIS is the problem ... you won't believe it even when it is true ...
Crime is down ... across eevery country & every US state & there is a very interesting & convincing explanation.
Because the crime-drop happens (approx) 18-20 years after another change ... the removal of Tetraethyl Lead from fuel.
Tetraetyhl is removed in 1980? Crime drops in 2000 ... Tetraethyl is removed in 1975? ... crime drops in 1995, or 1985/2005

etc ....

See also, thoughtful blog entry & LONG discussion HERE
PLEASE do read it, carefully.

Nick Drew said...

GT - is there any point in my remarking that I never once mentioned crime above ?

so - crime is dropping and this impacts on health tourism ... exactly how ?

Anonymous said...

"Whatever the stats say, when the welfare cost makes up the largest slice of the governmental spending pie, something is wrong"

The single largest item of spend is pensions. Followed by health. Welfare at number 3.
Even looking at the pie-chart just for lumped-together welfare (with tax credits (working and child) integrated) spending on the "idle unemployed" is only a tiny fraction of that on pensioners (state and ex-public servant). Of course, when you include other benefits that pensioners get as a result of low income, the pension "pot" gets even bigger.
You should never forget that when the state pension was started it paid to people who were supposed to retire some 20 years after they should have died !
Taking life expectancy we have today, the retirement age should be about 104 years.