Quality or quantity? - by Richard... Thursday, April 15, 2010

Quality or quantity? - by Richard... Thursday, April 15, 2010

"Voters, offered a list of policies, can't tell which party is putting them forward. People think the national insurance cut is a Labour plan, and that restrictions on company takeovers are being proposed by the Tories. That's a sign of how little attention most people pay to politics and how much of their reaction to it is shaped by emotion and instinct, not rational analysis."

This is an offering from Julian Glover in The Guardian on the nature of politics and the difficulties in changing peoples' minds. Commenting on the difference between the 2005 election result and the current polls, he observes that:

... all those speeches, conferences, rows, a boom and a huge bust, the end of one war and the worsening of another, three Lib Dem leaders and two each for the Tories and Labour – have shifted just three percent of voters from one party to the other. It's a chilling fact for anyone (such as me) who has spent those five years following and writing about politics in sometimes hyperbolic terms.

What does not seem to have occurred to Mr Glover is that changing minds might have something to do with the quality of the words – or the ideas behind them, rather than the quantity. And if people are mixing up policies, since all the parties look and sound so drearily similar, is it a wonder that people can't tell them apart?

Then to dismiss peoples' reactions as "shaped by emotion and instinct" then comes over as a tad patronising. I see more "rational analysis" out on the streets than I do either in the media or in the political élites.

GENERAL ELECTION THREAD

Print ready

Oxburgh – not a whitewash - by Richard... Wednesday, April 14, 2010

Although seen and widely reported as once again exonerating the CRU scientists at the centre of the "Climategate" scandal, it is important to note that the so-called Oxburgh report, published today, had a very limited brief.

The essence of the criticism that the Panel was asked to address – or so it tells us - was whether climatic data had been "dishonestly selected, manipulated and/or presented to arrive at pre-determined conclusions that were not compatible with a fair interpretation of the original data." 

The Panel, we are reminded, was not concerned with "the question of whether the conclusions of the published research were correct." Rather "it was asked to come to a view on the integrity of the Unit's research and whether as far as could be determined the conclusions represented an honest and scientifically justified interpretation of the data."

With such a limited remit, its substantive conclusion was that, it "saw no evidence of any deliberate scientific malpractice in any of the work of the Climatic Research Unit". Offering more than a minor hostage to fortune, it then goes on to say that: "had it been there we believe that it is likely that we would have detected it." 

Rather, the Panel concludes, "we found a small group of dedicated if slightly disorganised researchers who were ill-prepared for being the focus of public attention. As with many small research groups their internal procedures were rather informal."

In response, it comes as little surprise to find that Steve McIntyre is rather less than impressed, and in particular for one very good reason.

Perversely, Oxburgh and his merry men, in their rush to give the CRU a clean bill of health, make an entirely false distinction between the unit's output – which they say drew attention to the limitations of and uncertainties in their work – and the work by the IPCC and others which "have sometimes neglected to highlight this issue."

What McIntyre quite rightly points out is that, in terms of the temperature records produced, the CRU and the IPCC were effectively one and the same – the responsibility for the IPCC's output rested largely with Briffa and Jones. They, along with Michael Mann, actively prevented the limitations of and uncertainties in their work being included in the IPCC reports.

This, more than anything, is what is seen to turn the Oxburgh report into what could be called a "whitewash", although a commenter in The Daily Telegraph offers the words: "systemic political corruption". 

On the face of it, it is hard to disagree with this. As McIntyre notes, the Panel was announced on 22 March and its report is dated 12 April – three weeks end-to-end. They took no evidence and their list of references is 11 CRU papers, five on tree rings, six on CRUTEM. Notably missing are the 1000-year reconstructions, which must count as the most controversial and influential pieces of work produced by the unit.

Had the Panel been serious in its endeavour, it would most certainly have interviewed (or at least communicated with) McIntyre, and possibly Andrew Montford, of Bishop Hill fame. But it did neither. In fact, all it did was interview members of the CRU. On these grounds, the Panel fails and fails dismally.

However, while the failures are obvious and manifold, there is something of the Curates's egg in the report. For instance, there are the references to the equivocal nature of the science. In selecting the appropriate data, "a great deal of judgement has to be used", says the Panel, what to use and discard are "all matters of experience and judgement" and "the potential for misleading results arising from selection bias is very great in this area."

These comments need to be put in context with the second of the Panel's conclusions. Surprise is expressed that research in an area that depends so heavily on statistical methods was not carried out in close collaboration with professional statisticians. 

There would be mutual benefit, the Panel says, if there were closer collaboration and interaction between CRU and a much wider scientific group outside the relatively small international circle of temperature specialists.

Such is the nearest thing one finds to real criticism, but the lines are broad and reading between them suggests the recognition of a major failure in the use of statistical techniques.

Thus, we can put together a picture of the selection of data requiring "a great deal of judgement" which is extremely prone to selection bias, which was then subject to statistical techniques without the input from professional statisticians. And on top of that, there is an acknowledgement that judgemental decisions made had not been properly recorded, so that the work could be replicated by others.

Even in the limited terms of this report, therefore, we have serious doubt cast upon the adequacy of the CRU's work, with the only substantive finding being that any misrepresentation was not "deliberate". Whether it was or not, there is enough there to suggest that this is not work on which it would be safe to base policy decisions which are set to cost trillions of dollars and caused major economic disruptions.

Alarmingly, though, while the egregious Louise Gray in her report in The Daily Telegraph does not see it that way. 

She is forced to concede that the scientists had been criticised for being "naive" and "disorganised", but then goes on to say that there was no evidence of "deliberate scientific malpractice". From this, she concludes (her words) that "the conclusion that mankind is causing global warming is probably correct."

Not by any stretch of the imagination can this be supported, especially as the Panel – as we saw – set out to exclude from its brief "the question of whether the conclusions of the published research were correct." Grey has crossed the line and gone well over it, injecting falsehood into a news report.

However, Gray is not alone. The Nature blog headlines: "science solid despite lack of statistical know-how", which again cannot be imputed from the report.

At least though, the Telegraph leader notes that the findings of the Oxburgh inquiry are not an excuse for again closing down the climate-change debate to the exclusion of those who take a sceptical attitude to what is arguably the most important issue facing the world.

That much is true, but much more could and should be read into the report. Thus to brand it simply as a "whitewash" and to dismiss it entirely on those grounds would be a disservice to the sceptic cause. There is much that is useable in the report which, although muted, is nevertheless, an indictment of the CRU and climate science. To that extent, it is not a whitewash.

COMMENT THREAD - CLIMATE CHANGE

Print ready

The plague party wins it - by Richard...

Politics Home is running its own analysis of the election and iscurrently predicting an overall majority of 16 for Cameron (pictured).

The methodology, however, looks unconvincing but, more to the point, gut instinct says the prediction is wrong. What looks closer to reality is the latest Populus poll for The Times, which puts Labour on 33 and the Tories on 36 percent, narrowing the Tory lead to three points. The Lib-Dims are on 21 percent, so Labour would be the largest single party, with about 300 MPs, ahead of the Tories on 264 and the Lib-Dims on 54.

The paper headlines this as the general election "race" tightening, but that is probably to miscast it. People are walking away in their droves, reacting against the saturation coverage and the sight of a political class that seems to have learned no lessons, and has failed to realise quite how much they are loathed.

Thus, the subsidiary "take" of The Times is probably as real as it gets. The poll, it says, reveals deep public disenchantment with the campaign so far, with more voters now hoping for a hung parliament than either a Tory or a Labour outright victory. Some 32 percent of the public are in that camp, compared with against 28 percent wanting a Tory majority and 22 percent a victory for Labour.

There is talk of Labour enjoying a manifesto "bounce" and hence the Tory position is expected to recover in the next poll, when their manifesto launch registers with the public. But, while some Tory commentators are gushing over the little purple book, my hunch is that any bounce will be short-lived.

This manifesto, on reflection, is possibly the biggest political mistake Cameron has ever made. Too clever by half, it lacks in credibility and ignores the "elephant in the room" of the EU. And if people don't buy into it wholesale, they will most likely take the view that The Boy is trying to pull a fast one. There will be nothing in the middle.

Still the game is for playing, but that gut feeling says The Boy has blown it. A serious manifesto spelling out the problems and the remedies needed could have been a game-changer, but this thing is too vague and too gimmicky to do otherwise than suggest Cameron is playing games.

So, we end up – according to the current poll – with two fifths of the public (42 percent, up one point on a week ago) saying it wants a change from Labour, but are not sure that it is time for a change to the Conservatives. Some 33 percent, down one point, think it is time for a change to the Tories. The "plague on both your houses" party is winning this election.

GENERAL ELECTION THREAD