Page 7 of 8

Re: Lies, damned lies and statistics

Posted: 18 Jun 2015, 17:44
by plaques
Piketty and Stiglitz are trying their best to point out that America is the country with the highest levels of inequality followed by the UK. It looks like Osborne is trying to get us into the No 1 slot. Of course the Republicans will do their damnedest to keep first place. Osborne will have a fight on his hands. But with 5 years to go pity the poor sods at the bottom.

Re: Lies, damned lies and statistics

Posted: 19 Jun 2015, 04:23
by Stanley
Depressing isn't it. Richard is absolutely right (as usual....) and I see exactly the same picture. My head keeps going back to ;Healthy or Hungry Thirties' (on the site) and Naomi Klein's book, 'The Shock Doctrine'. I have this nagging thought that we are being subjected to a huge con trick by which the distribution of wealth and opportunity is gradually shifted up towards the wealthiest and the lower classes are made more vulnerable and open to control. All in the name of pursing 'sound money'. Society is being engineered by the right wing and we are sleepwalking into it. Or am I just a paranoid Old Marxist Fart?

Re: Lies, damned lies and statistics

Posted: 19 Jun 2015, 09:41
by Tizer
The Danes have just ousted their centre left party and elected an anti-immigration party. A commentator was saying that this will please Cameron because he'll get more support now.

Re: Lies, damned lies and statistics

Posted: 20 Jun 2015, 04:42
by Stanley
The bottom line is that (Particularly in the cases of immigration and austerity) governments are making such a mess of management that the electorate is rebelling. Expect more of it. Doesn't hold up much hope for political agreement in the EU over integrating budgets and governments to create the federal system which is the only way out to make the Euro work.

Re: Lies, damned lies and statistics

Posted: 21 Jun 2015, 05:36
by Stanley
I keep hearing the figure of £12billion annual cuts and £60billion over the five year length of the Parliament but I am not clear whether this is the total of all cuts or only welfare and public services. I wonder how often the goalposts will be moved as they fail to achieve what looks like impossible totals.

Re: Lies, damned lies and statistics

Posted: 21 Jun 2015, 08:29
by Tizer
Reuters, today:
"Writing in the Sunday Times, Osborne and welfare minister Iain Duncan Smith said the newly re-elected Conservative government would go ahead with plans to cut welfare spending by 12 billion pounds, out of an annual budget of 220 billion pounds, including old-age pensions." LINK

Re: Lies, damned lies and statistics

Posted: 22 Jun 2015, 03:51
by Stanley
Lovely caring way to save money...... I didn't vote for them!

Re: Lies, damned lies and statistics

Posted: 24 Jun 2015, 18:07
by plaques
In his latest speech on child poverty he finally admitted that incomes had been falling. Obviously this was an oversight on his part because he was trying to prove that the changes in child poverty numbers had nothing to do with real poverty but were some statistical aberration caused by changes in earnings. But to quote Cameron:- 'because if you look at the recession, the measured rate of child poverty fell because, actually, overall people’s income in the UK was falling' . He then goes on to try to define the root causes. :-' by recognising the root causes of poverty: entrenched worklessness, family breakdown, problem debt, and drug and alcohol dependency”. So to take up Brufs point on mental health in an earlier posting, child poverty is down to "Scroungers", "Alchies" and "Druggies". So us god fearing people can safely put them in a box and forget about them. The one thing I must admit to not knowing was that the 2.5% rise in my pension caused a rise in child poverty. from 2.3M to 2.5M. ie: 8% rise. This is surprising since OAP's only comprise of 17% of the population as against round 83% if you add OAP's and Working Population together. Even if the working population got zero rises and the earnings they did receive were the same as the OAP, (a total nonsense), the OAP's would only contribute a relative 0.5% in total earnings. More smoke and mirrors but it must be right because Mr Cameron says so.

Re: Lies, damned lies and statistics

Posted: 25 Jun 2015, 03:58
by Stanley
Well done P. You have evidently been analysing the figures far better than me! To tell you the truth I can't be bothered because they are a minefield. Does anyone really believe that Cameron spends hours working on the stats? He is not like Godron who certainly did, he was obsessive. Every time Cameron quotes a figure it is one trotted out by his advisers in the brief they give him and I'm afraid they are strangers to the truth. Their job is to put a gloss on everything in favour of their masters.It's all futile of course, the truth will out eventually and one thing seems certain, things are not going to get any better as the next welfare cuts strike.... It's a pointer to just how bad things are getting under unrestrained austerity and how bad our governance is. These people are incompetent and stupid.

Re: Lies, damned lies and statistics

Posted: 01 Jul 2015, 05:39
by Stanley
The reports of the Greek default are fertile ground at the moment. Figures in the billions being tossed around like confetti but nobody clearly laying out where it all went.... My interpretation so far is that the majority went in interest payments to the banks and the bond holders.

Re: Lies, damned lies and statistics

Posted: 30 Jul 2015, 15:56
by Tizer
Anthony Reuben, Head of Statistics for BBC News, makes some observations on the current mis-use of survey data and of poor survey design: `Headline numbers: Problems with a huge survey' LINK

Re: Lies, damned lies and statistics

Posted: 31 Jul 2015, 02:58
by Stanley
All surveys and polls are open to doubt. This even applies to the Bank of England statistics.... Then there is the minefield of interpretation of the results....

Re: Lies, damned lies and statistics

Posted: 04 Aug 2015, 06:05
by Stanley
We are generally quite sceptical about statistics but I went and had a look at THIS and my first impression is that we can trust these figures on migration more than most. I see I was wrong about Belgium having the highest proportion, it's Luxembourg, I shall correct that.

Re: Lies, damned lies and statistics

Posted: 04 Aug 2015, 09:10
by Tizer
Having looked at that linked web page I conclude that your eyesight really is as good as you claim! :smile:

Re: Lies, damned lies and statistics

Posted: 05 Aug 2015, 03:59
by Stanley
Quite but big screen helps.... Some fascinating figures in there.....

Re: Lies, damned lies and statistics

Posted: 24 Sep 2015, 02:35
by Stanley
The programme last night on BBC2 about world poverty was a brilliant use of statistics. If you missed it it is well worth a trip to Iplayer....

Re: Lies, damned lies and statistics

Posted: 02 Mar 2016, 09:38
by Tizer
I've put this link to an article about smartphone security here rather than in the Mobile Phones thread because the interesting part is towards the end - the way we are easily fooled into thinking coincidences have something ulterior behind them. I particularly like what happened to the man who wrote a book about coincidence:
"Professor Hand is not immune to the lure of coincidence himself. When his book was published another author published a very similar title at the same time. The author of The Coincidence Authority, John Ironmonger, shared the same birthday as Prof Hand and was based at the same university as his wife. "These sorts of things happen," he said. "Just because I understand why it happened doesn't make it any less beautiful."
LINK

Re: Lies, damned lies and statistics

Posted: 03 Mar 2016, 05:35
by Stanley
Who was it said "Once is happenstance, twice is coincidence, three times is conspiracy" My head is saying Damon Runyan but I'm not sure. Here's a LINK to his biography on Wikipedia. I looked the quote up as well and they all attribute it mainly to Ian Fleming but I still think I first came across it in one of Damon Runyan's short stories.....

Re: Lies, damned lies and statistics

Posted: 15 Dec 2016, 09:35
by Tizer
Failure to check simple conversions of units is rampant in the news media. Such as this on the BBC web site today...
"Artist Luke Jerram says he's planning to bring his 'museum of the moon' back to Bristol next year. The 3ft (7m) in diameter globe was due to be one of the main attractions at this year's Balloon Fiesta - but had to be taken down because of a tear." (The photo accompanying this news quite obviously shows that the 7m has to be the correct figure.)

Another case of not checking figures is shown by the web site of the Minster company, part of Jewsons, which sells building products. It has pages showing all the different types and sizes of Celotex insulation panels. In every case it shows a figure for the U-value (lambda value, or thermal conductivity) that is 10 times greater than the true value - they've got the decimal place wrong. They've got 0.22 W/m.K but it should be 0.022 W/m.K. I emailed the marketing manager and he thanked me and said they would change it. I notice it's still the same on the site.

Re: Lies, damned lies and statistics

Posted: 15 Dec 2016, 17:49
by Pluggy
Fairly common problem, most people wouldn't have a clue what energy units are about anyway. Very few in the energy sales sector know the difference between a kW and a kWh, and they wouldn't know a MJ if it bit them on the xxxx.

Re: Lies, damned lies and statistics

Posted: 16 Dec 2016, 04:42
by Stanley
Some don't even know the difference between millions and billions..... (or they can't be arsed to get it right.....)

Re: Lies, damned lies and statistics

Posted: 12 Apr 2017, 10:09
by Tizer
Tizer wrote: 15 Dec 2016, 09:35 Another case of not checking figures is shown by the web site of the Minster company, part of Jewsons, which sells building products. It has pages showing all the different types and sizes of Celotex insulation panels. In every case it shows a figure for the U-value (lambda value, or thermal conductivity) that is 10 times greater than the true value - they've got the decimal place wrong. They've got 0.22 W/m.K but it should be 0.022 W/m.K. I emailed the marketing manager and he thanked me and said they would change it. I notice it's still the same on the site.
I'm glad to be able to report that Minster/Jewsons have now corrected the figures on their site to show the correct values. Somebody has to keep an eye on these companies!

When I started this thread I didn't realise we would soon move into a world were lies and misinformation became the norm. Now they're everywhere and are undermining our trust in institutions and politicians worldwide. On a lower level misinformation insinuates itself into everyday affairs. I notice today that the National Osteoporosis Society is telling people to drink more milk and is claiming that skimmed milk contains more calcium than full fat milk. That's a fact, yes a genuine, real fact - but in nutritional terms there is no significant difference. The content varies but a typical difference is skimmed milk 244mg/serving, full-fat 236mg. In fact the variation between samples of skimmed will vary by more than that difference. My guess is that the Society is peddling the same old story of avoid dairy fat, avoid saturated fat.

We now have to beware of not just of lies but of facts which are genuine but are being used inappropriately to mislead us. Politicians, campaigners, marketing and sales people are all masters of these Dark Arts! :smile:

Re: Lies, damned lies and statistics

Posted: 13 Apr 2017, 03:47
by Stanley
I agree Tiz. I know it's boring but I always go back to Charles Webster and 'Healthy or Hungry Thirties'. Sometimes wrong statistics are simple error but in too many cases are the result of partial selection using only the 'facts' that support the argument. In the Thirties this was done by only quoting regional and national average figures and not recognising hot-spots where they were at their worst. This is a common ploy today especially in health. The present concentration of attention on Shrewsbury and the dead babies is a good example. This was a hot-spot of the worst kind but slipped though by not showing up in the national averages. Only yesterday the Trust was issuing statements saying that, overall, they compared well with this average. When pressed the man in charge descended into 'management speak' which is the next standard line of defence. Think of the number of times, when a politician is faced with an awkward question about funding they default to the formula "we are spending £Xbillion more", leaving aside the relationship of this to increased expense, often down to government polices.
Note also that in the case of Shrewsbury one brave family has fought to get the Trust to admit and modify their procedures for seven long years after their baby was a victim. All they ever got was reassurance that 'lessons have been learned'. There has never been a time when citizen scrutiny and protest has been more necessary.

Re: Lies, damned lies and statistics

Posted: 06 Jan 2018, 09:58
by Tizer
I was puzzled by a Guardian front page this morning. It has a story titled `Phones fuel sharp rise in exam cheating' which is accompanied by a big coloured circle containing the figures `50%' and below it the caption `The percentage of students found cheating by using unauthorised materials.' That sounds extremely high and this figure doesn't seem to be supported by the text front page text.
Here is the article on their web page: LINK
That page states: `Unauthorised materials accounted for half of all students given penalties for cheating, and of those nearly 80% related to the use of mobile phones.' This shows how their front page `50% ' graphic and caption is ambiguous and likely to mislead.

Re: Lies, damned lies and statistics

Posted: 06 Jan 2018, 10:24
by PanBiker
I have invigilated dozens of exams now and never experienced anyone using a mobile to cheat. Warning notices cover all electronic devices such as phones, smart watches, fit bits and any other electronic ephemera. Students are required to surrender mobiles on entering the examination room or have them switched off in their bags, they are now required to remove watches, they can keep these in view on the desk, wristbands of any type are not allowed. We have had one mobile ringing in a bag where the student has forgotten to turn it off (bags are deposited away from the students). Water is allowed as long as the label is removed from the bottle, i.e a clear container,directly printed on or coloured bottles are not allowed, any brought are confiscated. Students are given every opportunity to surrender contraband before the examination commences. It's down to the organisation and invigilation regime in the schools. All incidents have to be logged and signed off. Any with failure figures as quoted must be pretty slack set up and would fail at inspection when the logs are examined. All schools are visited by an inspector at some point during the exam schedule.