Showing posts with label Statistics. Show all posts
Showing posts with label Statistics. Show all posts

Sunday, 11 December 2011

Warning: Creeping Stateism is bad for your Liberties.

A “>report has been published in the British Journal of Cancer. It concerns the possible causes of various, predictably enough, cancers. Fair enough and all well and good so far.

It is based on medical researchers doing correlations and statistical analysis rather than any clinical tests, that is always prone to possible mis-interpretation, but it is not unreasonable to take the findings at face value.

It finds that lifestyle choices influence your likelihood of getting certain forms of cancer. That smoking is the main cause of lung cancer and second but far behind is a lack of fruit and vegetables in the diet possibly responsible for oesophageal or gullet cancer, half of the risk comes from eating too little fruit and veg. too much salt in the diet a possible cause of Stomach cancer. Way down there under 5% is drinking too much alcohol and being overweight.

Useful information if taken sensibly, along with other studies that show moderate levels of alcohol seem to be actually beneficial.

So what is apparently the knee jerk reaction of the Royal College of Physicians? True to form a demand for authoritarian legislation. Their president, Sir Richard Thompson, claimed the findings were a “wake-up call to the government” to take stronger action on public health.

He stated that rising incidence of preventable cancers showed that the 'carrot' approach of voluntary agreements with industry is not enough to prompt healthy behaviours, and needs to be replaced by the 'stick' approach of legislative solutions,"

Maybe Sir Richard Thompson’s intentions are good, but his instincts seem to be to order, to force and that is not.

Diane Abbott, Longstanding New Labour luminary and current Shadow Public Health Minister, said: "The government is failing on all the main public health issues.”
So let’s leave aside ideas of punitive taxation of burgers and sweets, making it illegal to smoke anywhere else or legislating illegal anti competitive minimum prices for alcohol for a moment. Lets rewind.

That reflexive authoritarian statist demand that the government get involved, do more.

Exactly what business of the state is it if I want a glass of wine after dinner in the fist place?

OK so I am not as fit or as skinny as I used to be but I can see how it is my business, maybe my “significant other”, my tailor even, but the Government?

The only justification I can see is how much I might cost the state. Possibly in increased pension payments from a scheme that I am basically forced by law to pay into by the Government if, I live significantly longer due to good lifestyle choices. Or increased medical costs from a scheme that I am basically forced by law to pay into, use it or not, by the Government.

So the only real justification for the State to be involved, apart possibly from the odd public health warning, is how much I cost systems that they force me to participate in and thus cost er - well me.

Catch 22.

Friday, 27 May 2011

Lies. Damned lies.. and 'health' statistics

The charity Alcohol Concern Cymru's (AAC) have apparently convinced themselves that there is drinking problem amongst Welsh seniors.

It conjures up an image of seniors staggering out of bars, vomiting in the street, getting into fights and sleeping it off in police cells.

AAC are alleging seniors drinking over the recommended maximum in the week before rose from 22% in 2003/4 to 34% in 2009 for men and 7% to 17% for women.

The charity's local manager Andrew Misell made the alarmist claim that: "With the number of people over retirement age increasing, some researchers have even talked about a silent epidemic of older people's alcohol issues."

Alcohol Concern Cymru's (AAC) is what is sometimes referred to a "fake" charity. That is, a registered charity that receives the bulk of it's funding, not from charitable donations, but instead from the state, or interested state bodies, often by quite convoluted paths.

In this case it receives funding from the Welsh Assembly and the Department of Health. One can't help but suspect the objectivity of such 'charities'. It is in their interests to talk up 'problems' and to dance the tune of their major funders.

So what are the limits alleged to be exceeded the week before? For men 21 per week and 14 for women. A unit is 1/2 pint of beer, a glass of wine or a shot measure.

It means a man who had more than a pint and a half each day in that week would exceed that limit as would a woman who drank more than a couple of glasses of wine each day would be accounted problem drinkers.

Presumably, whatever their general level of alcohol consumption is, it can’t have done them any great harm over the last 65 years, or they probably wouldn’t have got to be seniors in the first place.

One wonders when the survey was actually done. After the New Year? During the holiday season?

In any event, the big problem with the statistics (gathered by the Welsh Health Survey) that AAC used, is that the way things were measured changed in 2006 - and they did not take that into account. So by comparing 2003 with 2009 they are basically comparing apples to aardvarks.

AAC were making the mistake of only measuring the change in the way the figures were recorded and compiled - not in the way people drink at all. Oops…

Still it made a great alarmist headline didn’t it - and how many people will ever notice?

Wednesday, 16 April 2008

Alarmist predictions of 5 foot rise in sea level

According to a report from a UK/Finnish group sea levels could rise by up to almost 5 ft (4 ft 11 ½ inches) by the end of the century.

But ‘up to’ clearly includes any figure below that, including no discernable difference at all.

Apparently the team has built a computer model that can reflect the relationship between temperatures and sea level over the past 2000 years.

For the model’s predictions outside normal parameters to work it has to be accurate outside the parameters. It is simply not possible to test it against reality without data and many a model that accurately reflects relatively chaotic behaviour within certain bounds fails singularly when taken outside them.

It is an fact that global temperatures were significantly higher than current levels, between the 9th and 14th centuries. A period of some 500 years when temperatures were warmer than those today, what does the model say about sea level and ice cover during that period?

There is also the fact that the data on temperature predictions fed into such a model has to be accurate for it to be accurate. GIGO as they say.

The fact is that the global temperature of 2007 was statistically the same as 2006 and 2005 and every year since 2001. Unless the IPCC does a Robert Mugabe on the figures ‘global warming’ appears to have, for the moment at least, halted .

What happens next is anybody's guess - and that’s really what it is, a guess.

It could warm up to something more like it was back in the middle ages, it could remain stable, conceivably it could drop. Whatever happens it will sooner or later change, one way or the other.

That’s what climate does - and has done since long before humans learned to harness fire.

Maybe politicians on the AGW bandwagon should think twice about bio fuels and punitive taxes before they cause a crisis of their own, a food crisis.

Mind that’s no reason not to cut pollution, or build nuclear power stations, or develop compressed air or hydrogen powered transport. That makes sense anyway.

Wednesday, 2 April 2008

Report claims home birthers at greater risk if transferred to hospital

Sometimes one wonders at what otherwise presumably professional people do with statistics.

Researchers writing in the journal British Journal of Obstetrics and Gynaecology (BJOG) examined all home births over a decade.

They apparently feel the study indicates there could be ‘serious risks’ for the baby when mothers who chose a home birth are transferred to hospital.

The implication being that home birth is the problem aspect.

The problem is the data is self selecting. They are examining births that have to be transferred to hospital because they have got into trouble serious enough to require it. Also they are not really comparing like for like.

As it stands the particular conclusion mentioned above can really amount to little more than a pointer to a possible direction for further research. Something as easily arrived at by educated logic. It is largely useless - and it needn’t have been, if those conducting it had planned more carefully and harvested the right data. One fears to some extent it is time and money not used to the best advantage it could have been.

Only around 15% of home birthers actually have problems that require they transfer to hospital, for second time mothers the risk is lower.

A valid comparison would have been to look at how women who elected to give birth in hospital, with no contra indications prior to giving birth, fared against women who elected to give birth at home with, no prior contra indications, fared.

One might expect to see that the possible wait and journey to hospital might disadvantage home birthers who got into trouble, compared to Hospital birthers who similarly got into trouble - but it is possible this could be offset by other factors, such as the risk posed by antibiotic resistant infections and other infections present in hospitals, or possible closer individual attention during the initial stages.

Whilst the actual report is more balanced, the BBC headline is dubious in that it baldly states “Home birth to ward increases risk”. It is actually misleading as this is by no means proven, or evident, even from the report. An accurate headline might have prefaced it with “Report claims”.

Tuesday, 4 March 2008

Average UK hospital waits rise under New Labour

Before New Labour came to power in 1997 waits for hospital treatment of more than 18 months were not rare, now no-one waits longer than six months. Hurrah! Yes? – No. Not necessarily.

Sounds good, at first sight though, dosn’t it? The sort of soundbite Gordon Brown would proudly boast of in PMQ, or at conference.

The thing is the average wait has risen from 41 days to 49 days. Only 8 days up you may think. But it is 8 days longer on average.

This is a symptom of New Labour thinking. Everyone must be brought to a level. It’s like education, rather than raise up those getting a worse service they bring down everybody to the lowest level. The lowest common denominator, as they used to say - when such terms were recognised and understood by more people. So practically speaking overall levels of delivery drop.

Just so with waiting lists. To drop that headline figure to 6 months it is true that the really long waits have been drastically cut – but so have the really short ones.

So why did there used to be really short waits?

The question is who prioritised those waits back then? Well Doctors did, based on medical need, the urgency of the case. Now they are prioritised in order to meet state targets. One size fits all. So if you really need a short wait…

As chairman of the British Medical Association's consultants committee, Jonathan Fielden, pointed out "Doctors have been stopped from using their clinical judgement and pushing people through the system when they need to be.”

So the question you need to ask yourself is: Are you actually getting a better service, or just taken for yet another ride…

Thursday, 7 February 2008

Study shows predisposition to ‘obesity’ is largely inherited

An authoratitive study indicates that so-called ‘obesity’ is strongly influenced by genetic predisposition. The figure they quote is 77%.

Interesting news that, based on past performance, the Government Health fascists will probably do their best to ignore, or claim proves the exact opposite of what it means.

Also it should be remembered the ‘obesity’ definitions were tweaked to take in more people. Increasing the numbers of officially overweight, or obese, in the US alone by 28 million over night.

This does not mean that, because you are predisposed towards being fatter you will actually be fatter - Just that you will probably have to work much harder at avoiding it than others.

Much what many people have been convinced of for years…

Monday, 4 February 2008

Winston Churchill and Sherlock Holmes. Many voters don’t know who was real and who wasn’t

According to a survey 47% of those surveyed thought Richard I (the Lionheart), was a myth. Worryingly 33% thought the same about Sir Winston Churchill and Florence Nightingale.

Conversely 59% though Sherlock Holmes and Dr Watson were real people.

Did the people conducting the survey especially look for thick people, or is it yet another indictment of State Education?

Possibly both - but more likely the latter, as even the those of the most modest mental ability, exposed to a half decent education would be more than likely to get such simple questions right.

More worrying, what else don’t the respondents have a clue about - and how does this affect important decisions they make in daily life, based apparently on completely mistaken information.

More worrying still ;-) These people can vote...

Wednesday, 30 January 2008

New Labour claim to be close to hitting MRSA targets

In 2004, the Health Secretary at the time, John Reid, set a target of slashing MRSA rates in half by March 2008, though the exact details were never exactly defined beyond the sound-bite level. Only last year the Department of Health felt this, even as ill defined as it was, was unattainable, according to a leaked memo.

Now it seems the Government are briefing the BBC that it will consider the infections target met if there are 963 cases, or less in the quarter April to June 2008.

There is more than one way to use statistics.

Recorded honestly they are like looking in the mirror when you get ready in the morning. They give you accurate feedback, allowing you to see if your hair is sticking up, or not - and allowing you to either ensure it is, or brush it down, depending on your taste.

Then there is the ’lies, damned lies and statistics’ way, so beloved of the State and developed to previously unheard of levels of sophistication under New Labour - where they fiddle what is measured and how it is measured, in order to bamboozle the voters into the impression that the Government have actually accomplished something.

This method will generally not work in industry when analysing sales (except occasionally at shareholder meetings - but not often, as shareholders tend to pay attention and remember as it is their money at risk), as it persistently results in disastrous decisions and the company in question going bust.

In the case of the State/public sector however, the positives generally outweigh the negatives - The reward for conning the voters is re-election, or advancement. Any losses can always be made up in increased taxation.

This might not necessarily be so if the Government were actually honourable ;-) or genuinely identified themselves with the citizen. If more of them were of ‘us’ instead of ‘them’, or if actually achieving something were more important than only appearing to do so, say because of an unavoidable reality check.

Monday, 12 November 2007

UK – Rubbish at Rubbish?

The Local Government Association (LGA) are reporting that the UK dumps more rubbish in landfill sites than any other country in the EU, despite massively increasing our recycling, reducing the amount of rubbish landfilled by 3.6% it seems European countries had also cut their landfill amounts.

I would be more impressed by these figures if I could find out the actual tonnage. As it’s easy to reduce the percentage of 100 things but 1,000,000 is much more difficult to influence.

Also It would be interesting to know exactly what these European paragons of rubbish disposal actually do with their rubbish when they are not using landfill as a means of disposal.

Wednesday, 7 November 2007

Site Stats and Browser Wars

Looking at the site stats it looks as if Firefox is a fraction ahead of MS IE7 at 44% and 41% respectively. IE6 is fading at 15%.

It would be interesting to know how this compares with other sites traffic. And if different sorts of sites get visitors who use browsers in different ratios.

Thursday, 25 October 2007

Site Stats

I have been looking at the CFG’s stats. Here are some of them:

Language Groups: 92% English, 2% Welsh.
Continents: 53% North America, 40% Europe, 5% Asia.
Countries: 47% US, 38% UK, 6% Canada.
Browsers: 53% Firefox, 24% IE6, 17% IE7, 4% Safari.

I have left out any categories showing 1%.

It is curious that there are more visits from North America than Europe. I note use of Firefox browsers is the same percentage as North American visitors, though I doubt all the former are from the latter. It is interesting to note the Firefox users are in the majority of visitors at the moment.

Tuesday, 16 October 2007

Lots more 20 mph limits, many to have ‘safety’ camera revenue raisers

The Parliamentary Advisory Council for Transport Safety (Pacts) want to institute many more 20 mph zones and are arguing for average speed cameras to be installed in them.

They say Local Authorities are keen to introduce them.

Not half (as they say), set to a hair trigger they could be a highly lucrative new source of stealth taxation on the motorist.

Apart from the desire to take your money, what is one of the main ‘drivers’ for this further state persecution of motorists?

What else but NU-Lab Government targets. It this case of reducing deaths and serious injuries on the roads. Sounds laudable enough, but don’t they all. I guess they don’t believe their own “If you hit me at 30 mph…” propaganda any more…

More to the point - exactly what proportion of those killed, or seriously injured, on the roads are actually in a 20 or 30 mph zone when it happens?

Also how exactly do the define ‘serious’? Is it the same sort of trick they use reckoning ‘problem’, or ‘binge’ drinker figures?

Lies, damned lies and politician’s statistics.

It would probably be even more effective if they made it law that any cars in the zone required a person walking in front of them with a red flag. They could make it compulsory to hire a, council provided and licensed, trained flag bearer. They could then hire out the unemployed as flag bearers a condition of collecting their benefits.

Several birds with one stone including a massive reduction in unemployment figures and the ability to tax all motorists using the area if they exceed the speed limit by one and a half miles an hour or not.

Let’s hope they don’t think of that…

Saturday, 1 September 2007

Researchers claim Sheffield's citizens are the luckiest in UK

Now I have nothing against Sheffield, but it seems a team of psychologists from Peterborough recon it’s occupants are the luckiest in the UK.

How do they come up with this amazing statistic?

Apparently the researches asked people how often they experienced ‘good luck’, like finding a £10 note and how often they experienced bad luck, like having a bird bomb them with droppings.

By the same criteria they judged those living in Plymouth to be the unluckiest in the UK.

Now is it just me? Or does it occur to anyone else that Plymouth must have a plentiful supply of seagulls, known for their tendency towards loose stools, whereas Sheffield is probably less well endowed in that respect.

Also maybe the inhabitants of Plymouth are less careless with their tenners, than the inhabitants of Sheffield must be, based on the survey evidence - there must presumably also be a looser for every tenner found in Sheffield.

So these researchers – They got paid to do this? That's more than lucky...

Thursday, 19 July 2007

Crime and the perception of crime

The UK Home Affairs select committee’s are concerned about how effective the police are. The acting chairman, David Winnick, said: "We know the police have had a major increase in funding over the past decade but it is much more difficult to tell what they have done with it." The committee said the number of officers actually rose by 11% in that time.

According to the British Crime Survey (BCS) (the credibility of which was called into question by the Smith Review of crime statistics), crime had fallen by a third from 1997 to 2006, whilst recorded violent crime had risen by 21%.

One has to question the official figures to some extent in any case. There is strong anecdotal evidence that the targets are actually driving police behaviour away from effective policing and towards producing results that look good on paper. Taking minor easy win offences, or in some cases non offences and turning them into multiple detections. It would also seem changes in the law and organisational practices are resulting in officers spending more instead of less time on paperwork.

Despite the official figures the public appear to be unconvinced.

It seems, according to the authoritative Prof Ken Pease and Prof Graham Farrell that the BCS underestimates figures by around 3 million offences per year by only counting crimes committed against repeat victims up to 5, so if for instance, you report your car has been vandalised 10 times in a year only half of them will count.

Nil desperandum, as they used to say in Rome, it seems ministers are now planning to launch a new strategy to ‘move public perception of crime’ into line with official falling figures…