Page 5 of 8

Re: Lies, damned lies and statistics

Posted: 10 May 2014, 05:18
by Stanley
I like the way Piketty handles statistics, entirely different from the politician's take. He always includes a rider to point out the possible omissions and variations in the figures and in consequence you end up with confidence in his assessment. Pity the politicians can't learn this skill!

Re: Lies, damned lies and statistics

Posted: 10 May 2014, 13:20
by Pluggy
Stanley wrote:Pity the politicians can't learn this skill!
If only it were about skill. :)

Re: Lies, damned lies and statistics

Posted: 11 May 2014, 06:43
by Stanley
You're right Pluggy. I know I'm an old romantic but I expect to see some evidence in my representatives that they think a bit deeper than short term party advantage. Unfortunately our system is structured in such a way that you can't 'get on' in the House unless you adhere to the party line, embrace short term party politics and stick rigidly to the party line.
Take our incumbent here in Pendle, when did you hear him voice an original thought or recognise that one of his tasks is to try to improve the lot of the disadvantaged. To read his pieces in the BET you would be forgiven for thinking that there weren't any. He evidently believes the old theory that 'a rising tide lifts all boats' that wealth 'trickles down' through the economy, long since proved to be wrong. He should read Stiglitz and Piketty and do a b it of thinking but like his fellow Tories, he won't and will simply follow Ossie in his paean of praise for the 'economic recovery'.

Re: Lies, damned lies and statistics

Posted: 15 May 2014, 08:53
by plaques
Has Ossie been caught out using dodgy statistics. Tax Avoidance Perhaps Pfizer will be another of these statistical aberrations.

Re: Lies, damned lies and statistics

Posted: 16 May 2014, 04:04
by Stanley
P, Ossie is just as guilty as the rest of them. He obfuscates on the 'crack-down' because despite his promises almost nothing concrete has been done about it. Look in Private Eye for the real story and also the enlightened strategies used by Ossie and his family. You don't bite the hand that feeds you.

Re: Lies, damned lies and statistics

Posted: 08 Jul 2014, 09:10
by Tizer
After Ossie it's now Ed Milliband who can't do his sums, saying that in the last four years the number of people waiting more than four hours at A&E had gone up from 353,000 to 939,000 and claiming: "That's an increase of 300%", when it's actually an increase of 166%. An increase of 100% would give 706,000; of 200% 1,059,000; and 300% 1,412,000. And this is the man who gained four A Levels including an A in maths and Bs in further maths and physics and then read Philosophy, Politics and Economics at Corpus Christi College, Oxford. Then went on to graduate from the London School of Economics with an Master of Science in Economics. It's no wonder that ordinary folk can't figure out things like % interest rates!

Re: Lies, damned lies and statistics

Posted: 09 Jul 2014, 04:39
by Stanley
It doesn't matter how many qualifications you have (or how few), these mistakes boil down to incompetence and carelessness. Did nobody on his staff (who doubtlessly researched the statistics) check for accuracy. Sloppy work, sloppy politics and is it any wonder we are in the state we are now?

Re: Lies, damned lies and statistics

Posted: 09 Jul 2014, 07:20
by Bruff
To be fair I don't think this is the most egregious example of a politician using figures to make a point or suit their own ends. The rise has been dramatic and linking the '3' to the '9' there to make '300%' for effect is perhaps good rhetoric but true, not actually correct.

There's no serious attempt to pull the wool over anyone's eyes here, in contrast to the 'food bank' stat favoured by the PM. That is, under New Labour food bank usage went up ten-fold, from around 4000 folk to 40000. True. And so this allowed the PM when challenged on the huge increases during this Parliament to ignore the increase and note simply that usage went up 10-fold under New Labour knowing that until the numbers hit 400 thousand, there'd been no ten-fold increase under his administration. Of course numbers have exploded through the 400 thousand mark now and so that stat's not used much, rather it is the contention that of course numbers are up as anyone would grab free food if they could (or words to that effect), which as we've noted tells you all you need to know about the Tory making it.

Had no idea Milliband E had maths, further maths and physics A'Levels - impressive. Further maths is notoriously difficult.

Richard Broughton

Re: Lies, damned lies and statistics

Posted: 09 Jul 2014, 08:42
by Tizer
That's the trouble, he's probably good at `notoriously difficult' maths but fails on the very simple maths. I've always found simple arithmetic easy but can't cope with difficult maths! We can all make mistakes and I can still remember stating on OG that the speed of light was about 186,000 miles an hour and Ian had to remind me that it's miles/second not mph! But making mistakes on OG is somewhat different to making them in Parliament.

Re: Lies, damned lies and statistics

Posted: 09 Jul 2014, 11:22
by Tripps
"Had no idea Milliband E had maths, further maths and physics A'Levels - impressive. Further maths is notoriously difficult."

I've also got that trio, and would have no problem at all with 7x8 =? Nor would I have had at any time since I was six years old. My reading of the situation is that the lad was put up to it by his dad, and the Chancellor had been briefed not to answer any arithmetic questions in any circumstances, for fear of embarrassment. He did demonstrate his skill at avoiding questions though, which is probably of greater value to him in office than simple arithmetic.


"

Re: Lies, damned lies and statistics

Posted: 09 Jul 2014, 16:46
by Bruff
Yes, and in addition answering the rather simple 7x8 brings the obvious retort of 'OK then clever clogs, what's the square root of 169?' Which might be a bit harder.

I don't have much of a problem with politicians not answering these sort of questions. I've more a problem them not answering, or agonising over, the more routine 'what's your favourite X?' for fear of it being interpreted for something or other. Mr G Brown famously did this when asked for his favourite biscuit.

Richard Broughton

Re: Lies, damned lies and statistics

Posted: 10 Jul 2014, 05:09
by Stanley
I think you are all right and simply coming at it from different directions. My personal beef is that the mind set that creates these speeches and PR hand outs does the electorate a disservice by assuming that they are all gullible and will swallow everything they are told. Another evidence of reliance of deference to politicians. After all, if they say it it must be true. (Mustn't it?)

Re: Lies, damned lies and statistics

Posted: 24 Sep 2014, 08:40
by Tizer
Big trouble brewing over `subjective presentation of DNA results' in trials and predictions which is expected to result in `murders and rapists attempting to have their convictions over-turned'. Earlier in the year our top forensic science expert warned that courts were given subjective summaries instead of the actual results.

Re: Lies, damned lies and statistics

Posted: 25 Sep 2014, 05:35
by Stanley
Cliché reply but true... Magic Bullets always have Achilles' Heels.

Re: Lies, damned lies and statistics

Posted: 16 Oct 2014, 08:52
by Bruff
There has been some discussion recently over a report suggesting banning smoking in public parks, so here's some stuff. I think it established that the increased risk of lung cancer from being in regular and close proximity to cigarette smoke (what we call passive smoking) is 20%. That sounds bad. But, the lifetime risk of lung cancer in a non-smoker is 1%. So what that '20% increase' translates as is a 1.2% risk in a non-smoker. Or an increase in the lifetime risk of 1 in 500. And this is regular and close proximity. Basically, in a park, any risk increase I wager will be so vanishingly small that any 'health improvements' would be massively offset by the cost of the policy's introduction and enforcement.

So it seems this is done on the basis of it being an 'offence to the senses', and if that is a route we wish to go down there are a number of offences I could name. The overly-perfumed anyone?

Oh and I'm not a smoker, though I used to be

Richard Broughton

Re: Lies, damned lies and statistics

Posted: 17 Oct 2014, 04:01
by Stanley
Good point Richard, any effects from 'passive smoking' have always been assessed in enclosed situations like a bar person working in a heavy smoking environment. It turned the walls yellow as well! My experience is that many people catch a whiff of my good tobacco smoke and it gives them pleasure because it smalls so nice and reminds them of their Father! Think Incense in RC churches and barbecue smoke, far more dangerous than a pensioner (and all pipe smokers are pensioners these days) sitting on a park bench. By the way, I hate the smell of cigarette smoke, it's the smell of burning paper and chemicals used to keep the fags alight.

Re: Lies, damned lies and statistics

Posted: 27 Oct 2014, 22:17
by Whyperion
Tripps wrote:"Had no idea Milliband E had maths, further maths and physics A'Levels - impressive. Further maths is notoriously difficult."

I've also got that trio, "
Can we start a club? (nil membership fees), FM difficult ? No Wonder I am only a minor genius- only scraped a pass grade. ( was reading I think one of the newer applied maths syllabus workings , does seem (AS level?) more comprehensive, relevant and time consuming than the 1980s A levels). The idea of doing FM and phys was because much syllabus overlaped - less to learn but learn it well, I got A/O electronics too under same reasoning, wish I had done statistics too ( but did it for 2yrs of degree and thesis on some elements of stats ). Because of the A Levels got put into extended maths groups at degree level (trust me to make a rod for own back on that - still wondering where the 10% loss from perfection from 1st year course went for me , the missing 55% for the economics stream was more understandable- I keep disagreeing with the textbooks ).

Re: Lies, damned lies and statistics

Posted: 09 Nov 2014, 10:07
by plaques
I’m beginning to think that the writers of the “Pendle Matters” went to the same school of economics as Osborne and Cameron. The recent piece on pension payments stretches the use of semantics and statistics to its limits.
Quote;- ‘Over the last four years the state pension has increased from £97.65 to £113.10 per week, giving pensioners £800 more each year.’
Clearly the total over four years has risen by £15.45 per week = £800 per year. But this is not the same as £800 EACH year. The next rise, assuming 2.5%, will give £2.82 per week. £148 for the full year. A far cry from the amount that is being implied. When will they learn that people are not stupid!

Re: Lies, damned lies and statistics

Posted: 10 Nov 2014, 05:49
by Stanley
P. one of the things that is quite obvious about the 'information' conveyed by Pendle matters and Bandwagon's weekly article in the Times/Leader series is that they are either straight regurgitations of Central Office communication guidelines, attempts to gain traction by citing support of motherhood and apple pie or dubious attempts to claim responsibility for initiatives that are actually the work of other people. I wonder how they are going to gain traction from Barlick's success in the High Street competition? I once managed to elicit a public response in the BET letters page from Andrew Stephenson that he would be 'honest and transparent. To the best of my knowledge he hasn't ventured into that forum again and certainly hasn't kept that promise. The question of the Noble Lord Ashworth's contributions towards the Pendle CP has never been satisfactorily explained.
They are sensitive about this still, it was they who first added the strap line 'Not paid for by taxpayers' to Pendle Matters. Now why did they feel they had to do that?

Re: Lies, damned lies and statistics

Posted: 10 Nov 2014, 12:40
by Tizer
The Treasury is using a cunning wheeze in it's press releases, choosing to base its data on RPI or CPI depending on which puts them in the best light:
http://www.bbc.co.uk/news/blogs-magazin ... r-29934599

Re: Lies, damned lies and statistics

Posted: 11 Nov 2014, 06:45
by Stanley
This lack of transparency is embedded as a feature of governance in the UK and is not new. Once again I recommend readers to look for 'Healthy or Hungry Thirties' which is an account of the manipulation of medical statistics right through the 1930s by the government's MOH. An impeccable source and a clear account of how the public was misled to reinforce the political point that even though the Depression was raging in the real world, public health was steadily improving. It was done by manipulating the statistics, not telling lies and this is exactly what is happening now.

Re: Lies, damned lies and statistics

Posted: 11 Nov 2014, 09:50
by Tizer
A nice example of a misleading news report today from the Daily Express whose front page headline shouts:
`Britain faces month of wild storms'
It begins, "Fierce gales and lashing rain are about to unleash a month of storm misery on Britain", and attributes this information to the Met Office 30-day forecast. The truth is that the Met report states:
“For November-December-January above-average UK-mean temperatures are more likely than below-average”
and
“Latest predictions for UK-mean precipitation favour near- or above-average rainfall for November-December-January”.
How do they get away with such gross exaggeration?

Re: Lies, damned lies and statistics

Posted: 12 Nov 2014, 06:09
by Stanley
Never let the facts get in the way of a good story.... Besides, what do they know about weather cocooned in their centrally heated offices in the Great Wen?

Re: Lies, damned lies and statistics

Posted: 03 Feb 2015, 07:54
by Stanley
Classic case of partial statistics a la Charlie Webster this morning. See THIS for a report by the committee dealing with broadband access where they quote a figure of 95% of the UK having a 'broadband' speed of over 2mbs. One MP points out that 18% of her constituency, mostly rural, is under this standard. Another case where a UK wide statistic is trotted out as the norm and not recognising that there are pockets of deprivation. Exactly the same ploy as the statistics on child poverty and infant mortality. 'Hungry or Healthy Thirties' makes exactly the same point. This is basic dishonesty and should be addressed.

Re: Lies, damned lies and statistics

Posted: 03 Feb 2015, 10:13
by plaques
Of course they confuse things even more by talking Megabytes. not megabits. "A total 850,000 people cannot access broadband at even basic speeds, equivalent to 2 megabytes per second." I get just over 6 megabits which is adequate for streaming.