Monday 28 February 2011

That Yougov poll and the missing question

You'll have seen the recent Yougov poll and you might even have thought about how the weighting affected it.  Like me, you may have been wondering how the raw data converted to the weighted data, how this population of party identifiers:
Labour 291
Conservative 222
Liberal Democrat 67
Scottish National Party 289
Others 67
None / Don't know 322

was adjusted to become this population in the weighted data:
Labour 478

Conservative 163
Liberal Democrat 126
Scottish National Party 201
Others 25
None / Don't know 266

Very crudely, that means that each person who identifies as SNP who was polled counted as 0.7 of a person and every person who identifies with Labour was counted as 1.6 (Con - 0.7, LD - 1.9).  Just as interesting is the newspaper weighting, though, this:
Express / Mail 196
Sun / Star 100
Mirror / Record 94
Guardian / Independent / Herald 219
FT / Times / Telegraph / Scotsman 188
Other Paper 146
No Paper 315
became this:
Express / Mail 151

Sun / Star 201
Mirror / Record 251
Guardian / Independent / Herald 76
FT / Times / Telegraph / Scotsman 75
Other Paper 252
No Paper 252

A Daily Record reader counts as two and three quarter people in this poll and a Sun reader as double while a Herald reader is trimmed to just one third of a person and a Scotsman reader to two-fifths of a person.  It becomes, perhaps, even more interesting when you read this piece by Mike Smithson where he points out that Yougov is using old circulation figures for its newspaper weighting (the Record's readership is now under 307,000) and the dangers inherent in that were already laid out by Nick Sparrow.

It's fascinating, of course, to speculate on how wrong the poll is, but that's not what most intrigues me.  What most intrigues me is that this poll was commissioned by the Greens and it had another question in it which hasn't been published.  The other question was how the people polled felt about the Greens' idea of increasing tax in Scotland to offset the effects of the Westminster cuts.  It's up to the client what questions are published and when - I can't help but wonder what on earth could have been in the answer to that question that persuaded the Greens that it should be kept secret.

Answers on a postcard to ...

Saturday 19 February 2011

The barbarians at the gate

Hugh Henry MSP, who has ascended to the giddy heights of doyen of the Labour group in the Scottish Parliament, lodged a motion on Thursday and it would appear that he is setting himself up as arbiter of all things cultural:
S3M-07964 Hugh Henry (Paisley South) (Scottish Labour): Creative Scotland Expenditure— That the Parliament is concerned about recently published details of Creative Scotland expenditure; considers that, in a time of austerity, Creative Scotland should consider grant applications more carefully before making awards; expresses concern at and cannot understand the justification for £58,000 of taxpayers’ money being used to fund a dance programme based on the works of Alfred Hitchcock, or paying for travel to Tonga to study Polynesian dance, and believes that ministers need to urgently investigate Creative Scotland’s spending and take action to show that taxpayers’ money is being used responsibly.
This great, towering figure of Labour's intelligentsia would appear to be starting his own war against poshlost devoid of armaments.  He will be planning, perhaps, an exhibition of Entartete Kunst, ignoring the quite clear connection between the progress of society and progress in the Arts, a progress that cannot be governed or directed by politicians without strangling the progress.  Artists of all kinds must be free (and should be encouraged) to prod politicians and governments with pointed art - Arthur Miller received funding from the US National Endowment for the Arts to write Death of a Salesman and four years later he was puncturing the McCarthy witch-hunts with The Crucible - The National Theatre of Scotland used public money to produce Black Watch, a play that was extremely critical of UK military involvement overseas - and public money is opening up opportunities for people from all kinds of places to make a mark

Politicians can't run art through our ideological prisms (although you'll have some difficulty in keeping politicians from claiming credit for successes) without destroying the essence that makes art capable of moving the spirit, of changing humanity, of touching the vitality of people.  That's why public bodies funding the arts have operated at arms length from Government - although it is interesting that Hugh Henry appears to agree with the approach of the Conservative / Lib Dem coalition which is imposing massive cuts and strange conditions on Arts Council England.  It seems that UK Ministers, like Hugh Henry, think that "a time of austerity" is a time to cut cultural funding and have forgotten (if they ever knew) that the genesis of the Arts Councils - and therefore Creative Scotland - was the bleak austerity of December 1939 when the intent was “to show publicly and unmistakably that the Government cares about the cultural life of the country. This country is supposed to be fighting for civilisation” which led to the creation of the fore-runner of the Arts Councils in 1940 - and this in the midst of a World War.  Robert Hewison puts it excellently:
The decision taken in 1940 that led to long-term funding of the arts was not taken on economic grounds, or for reasons of health, social inclusion or the prevention of crime. But it was a rational decision, based on a rational argument: that we are supposed to be fighting for civilisation. 
I am, of course, giving Mr Henry the benefit of massive doubt and assuming that he is merely labouring mightily with the concept of funding art without directing it.  He may be being wilfully ignorant or, worse, intent upon finding some spuriously populist cause celebre with no regard to the consequences of his actions.  

We fund art and do so politically blind because art is damaged when it is narrowed by politics and Hugh Henry's version of Socialist Realism would run the risk of damaging cultural advancement, of restricting and choking Scottish art, his ambition for Cultural Revolution is misplaced.  We may not like everything that is funded by Creative Scotland but no-one ever argued that everyone will like all of the art we see around us - I find Shakespeare's plays quite dull (a couple of his sonnets are ok, though), can't abide the work of Alasdair Gray (a heresy in these parts), and can't for the life of me understand what's so good about those Titians we're so collectively proud of; but I've delighted in some of the National Theatre's productions, I savour the works of Banks and of Bellany among others, and I like to spend the odd hour or two from time to time soaking up a gallery.

I may not be the person to decide where arts funding should go, but I'm prepared to bet that speculative prospecting in areas seldom trodden (say a dance programme based on the works of Alfred Hitchcock or paying for travel to Tonga to study Polynesian dance) is likely to produce nuggets of gold more often than walking the same old dusty streets.  Mr Henry seems to have a problem with modern dance as a performance art and I suggest that he takes some time to go and take some in, we've got dancers to be proud of in Scotland.

We may not be fighting a World War at the moment, but the campaign for our civilisation continues.  The barbarians are always at the gate and we should always be driving past them to improve ourselves.

Thursday 17 February 2011

Academics - a blessing and a curse

There are some disadvantages to having academics as friends - one of them is that they quite happily give you a good finger-wagging over the least wee thing.  Here's a message I received from a friend of mine who obviously likes statistics even more than I do.  I've taken her name off the bottom in case she hassles me again and I can't reproduce the graph properly, I can't make it clear (I'm sure I'll get pelters for that, too), I guess I just failed my exam!


I do not like the 2nd paragraph of your blog (well, what you are saying is fine but it is horribly covered in percentages). What you need is a nice pretty graph, like the one I have attached for you.

Also, if you don’t already have it (but you probably do and apologies if so), the basic formula for calculating a 95% confidence interval for any proportion in Excel is:
Lower 95% CI limit: =B2-(1.96*(SQRT(B2*(1-B2)/C1)))
Upper 95% CI limit: =B2+(1.96*(SQRT(B2*(1-B2)/C1)))
where (for example) ‘B2’ is the cell you’ve put the proportion of interest (e.g. Alex Salmond’s approval rating) and ‘C1’ is the cell you’ve put the sample size in (e.g. 1019).

When comparing proportions (for example Salmond’s approval vs Gray’s approval) if the confidence intervals overlap, they lie within the margin of error and there is no statistical difference between them. If they do not overlap then Salmond’s approval is likely to be significantly higher or lower (statistically) than Gray’s.

What you should see from the Mori data is that Salmond’s approval rating is significantly higher than that of all the other party leaders (and Tavish Scott’s is significantly lower), that there is no statistical difference in dissatisfaction towards any of the party leaders, and that significantly more people have an opinion on Alex Salmond than any of the other party leaders (and significantly fewer people have an opinion on Tavish Scott than either Salmond or Gray, but not Goldie).

NB: Should you like this formula and want to use it but in future for whatever reason you need a wider margin of error, you could take a 99% confidence interval by substituting the 1.96 figure for 2.58.

Don’t say I’m not good to you.

Wednesday 16 February 2011

Polling along in the blue ...

Mario Lanza couldn't have sung it any better. There's plenty of comment out there on the latest poll for May's election that shows the SNP ahead of Labour and it's all well worth reading with the indications being that the SNP lead, if maintained into the election, could result in an SNP Government second term.  It's still too close to call, of course, but there are encouraging signs.  The gender gap in SNP support is closing - it's not closed yet by any manner of means but it is closing - and the issues that people think are important are chiming with what we (the SNP) have been saying and continue to say.  Let's take a look at a couple of points, though, starting with one which the Peat Worrier touched on; the party leader approval ratings.

Salmond's rating is +16%, Gray's rating is -1%, Goldie's rating is +2%, and Scott's rating is -8%.  That will put a wee grin on the coupon of the First Minister but there's something else in there just as interesting - only 14% of respondents didn't have an opinion on Salmond's performance while 33% didn't know about Gray, 38% about Goldie and 40% about Scott.  Does that matter?  Well, my experience has been that everyone has an opinion on politicians they know about so I'll assume that 86% of the population sample know who the First Minister is (there's probably a few who do know who he is but haven't made up their minds yet), and reducing numbers know who the others are.  That's a massive assumption on my part, though, and may be wrong; but if I'm right you have the interesting point that more than half of the sample (51%) approve of Salmond and only a few don't know (14%), whereas 67% don't know about Gray or disapprove of him, 68% don't know or disapprove for Goldie, and 74% don't know or disapprove of Scott.  Is it important?  It may be - especially when electors are having a look at the runners and riders for Government and making a decision about which party is well led and which senior party members would make good Ministers in addition to choosing their local representative.  Makes it interesting, I think especially when you turn it round - Salmond has a 51% approval rating against a 35% disapproval, meaning that his approval numbers beat the approval numbers of all the other leaders, but his disapproval ratings beat their approval ratings as well as their approval ratings - still that in the bowl and mix it!

And so to tax - it and death, we are told are the only two certainties.  The tax question showed a clear lead for all income tax to be set and collected in Scotland with the second choice being all income tax set and collected by the Whitehall Government.  The least popular answer was the Calman Commission proposal with a figure which only just beats one in four, so I guess that's a short answer on that bizarre scheme:
I would prefer all income tax to be set and collected by the UK government as it is at present 32%
I would prefer some income tax to be set and collected by the UK government and some by the Scottish government 27%
I would prefer all income tax to be set and collected by the Scottish government 37%
Don’t know 4%

Here's another thing, though; the opinion poll is a massive bump for the SNP since the TNS poll published in the Herald in January, but the trends shown by Ipsos are also interesting, let's start with SNP v Labour.  Three recent polls, Aug 2010, Nov 2010, Feb 2011, and starting with the constituency vote:
SNP 34%, 31%, 37%
Labour 37%, 41%, 36%
that went from within the margin of error to quite far outside it and then snapped right back into it - neck and neck (although the SNP is now ahead by a thin margin).  The Conservatives and Lid Dems, meanwhile stayed within the margin of error but swapped places at the extremes of it which some might suggest is the Conservatives benefiting from their coalition at the expense of their coalition partners:
Con 11%, 13%, 13%
LD 13%, 11%, 10%

Look at the Additional Member vote, though, and a very clear pattern emerges:
SNP 29%, 32%, 35%
Lab 38%, 36%, 33%
When the electors are looking at the parties their approval is on the move from Labour to the SNP - it's a 5.5% swing over six months and the trendlines in those two supports is definite.  For the other two parties there isn't much news to report:
Con 12%, 12%, 13%
LD 12%, 9%, 10%
and the Greens are up from 5% to 6% - still not at the races.

What does it all mean?  Well, as Brian Taylor says in his piece, each opinion poll is only a snapshot, it's trends that are important, and it would seem that the trends are tending to favour the brave.  In the words of a former Leader of the Labour Party Group in the Scottish Parliament, bring it on!