Archive for the 'Pollsters/polling' Category


The best guide to GE15 will come from single constituency polls NOT the national surveys and the seat calculators

Tuesday, October 21st, 2014

At what level majority will Lord A find the Tories holding on in the marginals?

In the past year we’ve seen a revolution in British political polling which is totally transforming the way wrong look at General Elections.

Rather than the focus being on national polls from which we can project seat numbers we are seeing an avalanche of constituency polls coming mostly from Lord Ashcroft and initiatives funded by wealthy UKIP donors.

These are serious polls of single constituencies with proper sized samples and should not be compared with the marginals polling of yesteryear. In the Ashcroft ones the standard sample is 1,000 and we get both the specific seat data and the overall aggregate whenever a new round is published

So far we’ve had polls in just under 90 Westminster seats from Lord A, Survation and ICM. and there are said to be new ranges of constituency in the pipeline. Lord A gave us a taster in a recent post:-

“…Labour would become the largest party if results in the seats I have already polled turned into results on election day – and there could well be more to come: while my polling has moved into seats with bigger Tory majorities I have not yet come to the “bite point” at which the potential losses end and Conservative seats consistently start to stay blue.

Research I currently have in the field is looking at some of these safer seats in search of the point at which the damage stops. If and when we find it, that should define the boundary of the real Conservative-Labour battleground.

But other unknowns remain. For example, are there vulnerable but hitherto unpolled Lib Dem seats in England and Wales? Could UKIP be making a significant impact in places we have not yet looked at> And what is happening in Scotland, where the big SNP gains some expect could change the equation significantly, especially if they are at the expense of Labour?..”

I wonder how that bite point will compare with national polling when fed into a seat calculator. Based on what we’ve seen so far there’s a bigger CON-LAB swing in the battleground. In LD held seats it is hard to draw any conclusions. In some areas seat polling is going with national polling – in other areas it isn’t

As I keep on saying general elections are not determined by national aggregate vote shares but by 650 separate votes in individual seats and I for one am hugely grateful to Lord A for helping us see what is happening where it matters.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


Polling analysis: Rochester is a far far bigger challenge for UKIP than Clacton

Monday, October 20th, 2014

UKIP is not winning the 2010 Tory vote like it did in Clacton

Reckless has nothing like the personal support as Carswell

The outcome could be on a knife-edge

I’ve become totally absorbed by the Rochester by-election the outcome of which, either way, will have a dramatic affect on the political environment in the six months to the May 7th general election.

Over the weekend I’ve had a look again at the only poll so far which was from Survation. This had UKIP’s Mark Reckless with an 8.7% margin a large part of which was made up of non-voters from 2010 and a disproportionate number of those saying they voted for “others”.

    In fact if standard ICM methodology, rather than Survation’s, had been used with the same data then the main two protagonists could have been almost level pegging with Labour not far behind. This is because ICM discounts the views of non-voters from last time by 50% and also re-allocates part of the “will vote -won’t say” segment to the party they supported last. Also 2010 “others” would have been scaled down.

UKIP, of course, gave Reckless a free ride in 2010 so there’s no 2010 data relating to the party to link back to.

Lord Ashcroft, who hasn’t polled this yet, is much closer in his approach to ICM and when he does he’ll be naming the candidates in his survey.

Survation was first off with a Clacton poll and followed that up a fortnight ago with its Rochester survey. Apart from the voting ones questions were almost identical allowing us to compare the two sets of data to identify the differences.

The key ones to me are how much worse Reckless’s defection is viewed in Rochester compared with Carswell and how in Rochester the Tories are hanging on to much more of their 2010 vote. The comparisons are shown in the two charts and do not look good for UKIP.

I still think that Reckless is favourite but nothing like the 78% chance that he’s being rated at on Betfair.

Mike Smithson

Ranked in top 33 most influential over 50s on Twitter


When ComRes tested impact of prompting for UKIP the views of women barely changed. Male support however jumped by 8%

Sunday, October 19th, 2014

Two pollsters, three polls, and UKIP shares between 16% and 24%

With all eyes on UKIP polling shares following their by election successes the online survey by ComRes for the Indy on Sunday and Sunday Mirror carried out a test to see whether, as many purple enthusiasts argue, their shares are understated by firms that don’t specifically prompt for the party.

So the ComRes sample was split in two with the first using the conventional approach and the second including UKIP in its main party prompts.

The problem with this is that the sample sizes became so small, down to 782 in one case, that the margin of error increases substantially especially when trying to analyse the UKIP voting subset.

In fact the difference between the two approaches can almost all be explained as standard margin of error.

    With that caveat a big move was apparent between the two ComRes polls. The views of women barely changed when UKIP was prompted – men, however increased their support by 8%

Read into that what you will! Maybe prompting says more about how men and women respond to online polling than it does about UKIP support.

Another difference was that non-2010 voters amongst UKIP support amounted to 7% in normal poll, but 13% in the prompted one.

Meanwhile the latest YouGov, with a later fieldwork period than ComRes, has UKIP down 3% from dizzy heights of last week to a more normal looking 16%.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


Polling analysis: UKIP’s hurting CON even more in the marginals than it was 2 months ago

Friday, October 17th, 2014

Latest churn figures from main parties to Farage’s

One of the great things about the Lord Ashcroft marginals polling is the sheer scale of it and the size of the overall samples. He tends to operate with samples of 1,000 meaning that the latest batch involved talking on the phone to a total of 11,002 people which is the equivalent to almost a year’s worth of ICM or Ipsos-MORI polls.

The benefit is that the aggregate data from all the constituencies provides large enough sub-samples on which to do analysis and in this post I look at the breakdown of the UKIP vote. The data in the chart above is produced by taking the total number of UKIP voters and dividing that by the numbers who voted for CON, LAB and LD at GE10.

I did a similar exercise with Lord A’s August round when he was polling CON held seats with smaller majorities.

As can be seen far more 2010 CON voters in these battleground seats have switched to UKIP than 2010 LAB or LD ones.

The comparison between the two two of polling is even more pronounced with the percentage of CON>UKIP switchers in the UKIP total up by more than eight points. The LAB switching is up by nothing like the same scale. LD switching, meanwhile, drops a bit.

How’s this going to shake out on May 7th next year? We do see in this polling that when asked to think about their own seats some UKIP supporters switch to the main two parties but not that many. My reading is that the UKIP will decline because the high-octane campaigning by both the red and blue teams will present the fight as a choice between them.

In other less marginal seats I expect that UKIP will hold up far more.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


The polling’s not all good for UKIP: See this worrying data for Farage’s party from YouGov and Ipsos-MORI

Friday, October 17th, 2014

Last month Ipsos-MORI had them the most disliked & least liked party

Could we be seeing the basis for anti-purple tactical voting?

In a week that has been dominated by positive GE15 voting numbers for YouGov there’s some other data from firm for the Economist, see top panel, that might make uncomfortable reading. The way the party is perceived by a representative sample of voters.

Those numbers are not good for the party and raise the prospect, I’d suggest, of anti-UKIP tactical voting with people not supporting their allegiance but the party most able to beat Farage’s party. It was suggested that this might have happened in the Newark by-election in June.

Several people who were “on the ground” during that by-election have told me how they’d come across quite a level a “cross-over” voting for this purpose with ex-LD and even ex-LAB voters shifting to CON for the election to stop UKIP. We have seen this in the past where the BNP have been strong in a seat.

The conditions for this, I’d suggest, are where it looks likely that the purples might be in with a shout and where highly intensive ground campaigning is taking place increasing overall awareness of the election – Rochester on November 20th perhaps?

Much publicised surges can have their negative side.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


Ipsos-Mori becomes the third pollster in less than a week to show a record high for UKIP

Wednesday, October 15th, 2014

Following on from Survation and YouGov recording their highest ever shares for UKIP, Ipsos-Mori joins the party, and has their highest ever share for UKIP.

Labour will be relieved to be back in the lead, but as with other pollsters, we’re seeing some historically low shares for the Con and Lab combined.

Labour will be alarmed that Ed’s ratings continue to slide, and are perilously close to Nick Clegg’s ratings.

For me the most interesting finding from this polling was this

More of the British public now disagree that voting UKIP in a general election is a wasted vote than agree. Just under half (48%) disagree that a general election vote for UKIP is a wasted vote, compared with 41% who agree. This is a turnaround from earlier this year – last month, prior to the Clacton by-election, 50% thought a UKIP vote was a vote wasted and 41% disagreed; in May, some 57% thought voting UKIP was a wasted vote and just 33% disagreed.

So those expecting the UKIP to fade by the General Election maybe in for a surprise, as people won’t view their UKIP vote as a wasted vote.



UKIP hit a record 18% from YouGov in the latest daily poll

Wednesday, October 15th, 2014

The reverberations from Clacton continue

After the strong UKIP performances in the most recent Survation, Ashcrfot and ICM polls YouGov is reporting this morning that the party is on 18% – the highest ever figure from the firm.

Survation at the weekend had the party on 25% while ICM on Monday saw UKIP move up 5% to 14%. The same day the Ashcroft National Poll had the party on 19% which equalled the previous record share.

So there’s strong polling evidence now that UKIP has, as you’d expect, benefited enormously from the winning its first ever MP at Clacton last week and the coming so close in Heywood.

    The latest YouGov has CON 30%/LAB 34%/LD 8%/UKIP 18%

With the upcoming Rochester by-election on November 20th there’s a strong chance that the party can remain at high levels. Insurgent parties need publicity and by-election successes can keep them in the spotlight.

As I’ve said in previous posts Rochester is by far and away the most important by-election of recent times. Quite simply the Tories have to stop the Farage band-wagon here or else prospects for GE15 will look even bleaker. At the same time a failure by Mark Reckless to hold onto the seat with his new allegiance would be a severe set-back for UKIP.

For those, like me, who love by-elections Rochester and Strood is a contest to savour. My guess is that the next polling will be after the Tory primary has been completed and the survey can name the main candidates.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


The single issues that look most set to determine GE15 votes: ICM’s new approach to what’s salient

Tuesday, October 14th, 2014

The gaps between the 2010 LDs and UKIP voters are enormous

In its latest phone poll for the Guardian ICM takes a novel new approach to testing the salience of specific issues and the impact on voting.

As can be seen in the chart above the sample was asked to state the single most important issue that would influence their vote. For me the big surprise is that immigration is pipped for top place by the NHS and that the key personal financial areas of jobs, prices, and wages come in third.

The NHS, as has been widely reported, is an area that Lynton Croby doesn’t want the blue team to talk about.

For the chart I’ve separated off for special analysis the two key groups of wing voters since GE10 – those now saying UKIP and those who voted Lib Dem. Click on the separate tabs to see how they responded and you have a totally different pattern of what’s considered to be important.

Whilst immigration his 43% as top issue for UKIP voters just 7% of 2010 LDs say the same. Surprisingly considering their age profile UKIP supporters are far less concerned about the NHS and just 2% made dealing with the deficit their top priority.

    This all illustrates the strategic dilemma for for Tories. Going hard on immigration might be the right way of winning back some of the voters who’ve gone to UKIP but it is not going to cause many 2010 LD switchers to return to their allegiance.

In fact it might be that strong immigration messages from the blue team reinforces the views of 2010 LD switchers making the Tory challenge even greater.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble