Archive for the 'Pollsters/polling' Category

h1

Opinium poll sees UKIP up six

Saturday, August 16th, 2014

UKIP sees a six point surge with Opinium

 

The usual caveats apply, this is but one poll, we need to see other polling to see if this UKIP surge is occurring or not. My own thought is that, this is more a return to the status quo for Opinium with regards to UKIP, their last poll, a fortnight ago, had UKIP down to their lowest point since February 2013, and that didn’t feel right. This UKIP’s highest score with Opinium since May 2013.

This is only the eighth poll out 1,611 polls in this parliament to have the combined Con and Lab score to be 60 or below.

The Lib Dems will be delighted to be polling in double digits, with the pollster that has consistently given them some of their lowest scores in this parliament. The fact the Lib Dems are polling double digits is considered good news (by me at least), probably explains the dire polling situation the Lib Dems find themselves in.

Putting this poll into electoral calculus, the result would be a Labour majority of 36, and zero UKIP MPs, as first past the post, and UNS will theoretically disadvantage UKIP.

You can get odds of 20/1 of UKIP polling between 20% and 25% next year, and 5/4 on UKIP not win a seat next year.

TSE



h1

Are we suffering from polling overload?

Friday, August 15th, 2014

Are we getting too many polls?

I know that might be a churlish thing for us polling addicts to say but the below graph shows the comparison of the number of Westminster VI polls conducted by BPC pollsters, in July 2009 and July 2014.

We’ve gone from eight polls in July 2009, to forty-four in July 2014, even if we remove the YouGov daily tracker, the number of non You-Gov polls has increased from five to twenty in the same period.

These figures do not include the various marginal/seat specific polls we get, or the Scottish Independence referendum polls.

Some days, we’re getting the same number of polls, that we would only get at the business end of a general election campaign.

So what’s the downside of all this polling?

I suspect, because there’s a newer poll out very quickly, and we move on to that before we’ve given the previous poll the care and attention that it deserves. Parties and their supporters don’t like one poll (an outlier), there’ll be one along shortly that will be more favourable.

Politics, is often about momentum, for example the last two Guardian/ICM polls would have dominated PB and the wider political narrative for a few days or longer five years ago,  in 2014 within a few hours, there was a newer poll, and we moved on from the Guardian/ICM poll.

PBer Gin1138 named it #MegaPollingMonday the two occasions last month we had four Westminster VI opinion polls on a Monday. Tomorrow night is a perfect example of what I’m talking about, I’m anticipating three Westminster Voting Intention polls, and at least two Indyref polls, so people will pick and choose the poll the suits them the best.

Polling maybe the epitome of less being more.

TSE



h1

Unique all female sample IndyRef poll finds 20% lead for NO

Thursday, August 14th, 2014

The front page of Thursday’s Daily Record is dominated by a new Survation poll confined to Scottish women only. This is the first time such a survey has been carried out and the figures, after the exclusion of the don’t knows, show a 20% NO lead.

Because the poll is unique we have little to compare it with apart from the gender splits in previous Survation polls.

It has been clear for sometime that the biggest hurdle for those campaigning for Scottish independence has been women who have generally been much more negative about the change than men.

My reaction on looking at the raw numbers is that this isn’t quite as bad as it might have been for YES. But still the message comes over strong as can be seen by the way the paper is treating it.

The part of the poll which the Record is highlighting are the views of the sample of Salmond, Sturgeon and Darling who is leading “Better Together”. 55% described the First Minister as arrogant compared with 31% for Darling.

Although the notion of a women only poll is interesting it would have been good to have some male data as well for comparison. Whatever an interesting survey from the innovative Survation firm and more bad news for YES.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble




h1

Ipsos-MORI finds CON and LAB level pegging – but with Boris as leader they’d be 5% ahead

Wednesday, August 13th, 2014



h1

Why LAB wins more seats with fewer votes : The way First Past the Post works in its favour

Wednesday, August 13th, 2014

Understanding Labour’s “other crutch”

We’ve talked a lot on PB about Labour’s “electoral crutch” – the big shift to it since 2010 of Lib Dem voters which has so far remained. Well Labour has another crutch – the electoral system which could be equally or even more important.

UK general elections are not decided by aggregate national vote shares but by FPTP elections in 650 separate seats where voters choose which individuals they want as their MPs.

Unlike the Euro elections the process is not about voting for parties but for people and whoever tops the poll in each of the 650 goes to Westminster. In one seat last time, Norwich South, the Lib Dem candidate won with just 29% of the votes.

Generally the party that chalks up the biggest aggregate national vote share ends up “winning” but not always. LAB won on votes in 1951 but the Tories were returned with a workable majority. In February 1974 Harold Wilson’s LAB secured fewer votes than Heath’s Tories but won more seats.

    The biggest driver of the seats:votes ratio is not as commonly believed the “boundaries” but the fact that LAB seats on average have significantly lower turnout levels than CON ones

The chart above shows the gap. Boundaries do play a part as the third drop down chart shows but not on the same scale as turnout. Added complications are that the Tories see many more votes “wasted” in seats where they come third and are much more vulnerable than LAB to tactical voting.

In what could be tight election on national votes shares Labour could easily repeat February 1974 and win on seats but lose on votes. If the tactical Anti-CON element is strong, which I believe it will be, then we could be heading for what could appear a perverse and unfair result.

It might just be possible that there could be a LAB majority on fewer votes.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble




h1

If the “ICM August polls before general elections” rule works again then Ed Miliband is home and dry

Tuesday, August 12th, 2014

Look at the record for when LAB is in the lead

Five years ago I made a comment on PB that August polls should not be trusted because of the holiday effect and got into an email exchange with Nick Sparrow – then head of polling for ICM.

He pointed out the following from his firm’s record over several general elections which on the face of it looks convincing.

August 1996 ICM poll had LAB ahead by 12%.
The result – LAB won by 13%

August 2000 ICM poll had LAB ahead by 10%
The result – LAB won by 9%

August 2004 ICM poll had LAB ahead by 3%
The result – LAB won by 3%

The only problem with this “rule” is that it doesn’t seem to apply with the Tories. The August 2009 ICM poll had Cameron’s party leading by a massive 16% ahead looking all set for a big majority. Their vote margin in the election itself was 7%.

There is a big point about this which Oxford’s Dr Stephen Fisher should note. Following the polling debacle at GE1992 ICM’s Nick Sparrow led the way in developing a different approach to polling which didn’t grossly overstate LAB. He was proved right at GE1997 and GE2001 and by the time GE2005 came along other firms had followed the Sparrow approach.

Much of Fisher’s polling trend data embraces the big LAB poll over-staters of the past.

Last night, as reported on the last thread, ICM’s phone poll for the Guardian had CON 31, LAB 38, LD 12 and UKIP 10.

Mike Smithson

Ranked in top 33 most influential over 50s on Twitter




h1

The August ICM poll sees reverse cross-over with LAB moving from a 1% deficit to a 7% lead

Monday, August 11th, 2014

And the inevitable “what if Boris was CON leader” questions

Tonight’s big polling news is that Labour has moved up sharply in the monthly ICM phone poll for the Guardian. In July EdM’s party was 1% behind. Now they are 7% ahead.

I must admit that I’m rather surprised by these latest findings and was expecting both main parties to be just about level-pegging. We’ll have to wait till the full data is out before we can work out what’s happened.

ICM tends to find UKIP with lower numbers than other firms partly because of its turnout filtering which scales back by 50% the value of responses of non-voters from GE2010.

Because of ICM’s long-standing reputation as the “gold standard” its finding are taken more seriously than just about any other pollster and inevitably tonight’s numbers will hearten EdM’s team as they go into next month’s conference season.

There were the inevitable BoJo questions with the inevitable findings. The LAB lead drops to 3% with the Mayor as the leader and he’s by far and away the most popular choice as Dave’s replacement.

The real problem for Cameron is that time is running out for the Tories.

Mike Smithson

Ranked in top 33 most influential over 50s on Twitter




h1

Another cross-over gets reversed – but could ICM, expected tonight, have a consecutive CON lead?

Monday, August 11th, 2014

This could be a highly unusual polling day for the two firms to be reporting both had the Tories in the lead last time out. A couple of hours ago Populus online moved from a 1% CON lead on Friday to a 4% LAB one now.

Later in the day I’m hoping that we should see the ICM Guardian poll which in July had CON 34. LAB 33, LD 12, UKIP 9. I’ve not had confirmation yet that this will be out but this is the normal point n the month when it appears.

Over the past four months we have see a handful of cross-overs but none have been sustained to the next poll from the same pollster. Could it just happen with ICM?

And how are we going to define a cross-over?

Mike Smithson

Ranked in top 33 most influential over 50s on Twitter