The mystery of the second ICM poll that was carried out last weekend

February 18th, 2015

Does it show similar figures to the published poll?

With much of the current GE15 narrative being linked to ICM’s 4% lead poll from Monday it is interesting to note that this was not the only survey being carried out by the firm at the weekend.

On Sunday, as I have reported, I was polled by ICM. The call was initiated from the firm’s big political calling centre in the Bromham Road in Bedford which is, incidentally, only about a mile from where I live. I know that because I asked the interviewer. Recalling the detail of the interview it does not match the Guardian poll published on Monday night. Having gone through the dataset from that survey it is clear that this wasn’t the series of questions that I answered.

Unusually for a political poll the main voting intention question was not put first. Instead the initial part of the survey points asked me to give ratings on a scale of zero to 100 my views of the main political parties and later their leaders.

Interestingly when it came to the voting question the prompt included a full range of parties including the BNP and the Greens as well as the normal big 4. It was as though that one of the objectives of the poll to specifically work out levels of support the parties that are usually classified as “others”.

There was questioning on the issues that I considered to be most important asking me to name two.

So who could have carried out this survey? My guess is that it possibly was the Conservative party itself. On reflection the poll appeared to have been designed to test how the party race was getting on and how effectively the blue team’s messages were getting across at this stage in the campaign.

Phone polls like this one don’t come cheaply. They’re not the sort of thing that a party or organisation that’s strapped for resources can be commissioning. My feeling is that it must have been linked to the Conservative machine.

    If I am right then the person who knows better than anybody whether the ICM Guardian poll was an outlier or not is Lynton Crosby.

Meanwhile we know from the other polls that were carried out at the same time and later that there has yet to be corroboration for the CON surge that ICM found.

Polling update

Mike Smithson

For 11 years viewing politics from OUTSIDE the Westminster bubble


As pollsters start to look at tacticals Marf gives her take.

February 17th, 2015

  • If you would like to purchase one of Marf’s prints or originals, please contact her here.

  • h1

    GE2015 comes down to Brand “Labour” versus Brand “Cameron”. Which will prevail?

    February 17th, 2015

    Meanwhile the negatives build up against the purples

    it has been said often in recent months that what the country really wants is a Labour government led by David Cameron. This perception is reinforced by the latest positive/negative in this week’s Ashcroft Nationaal Pooll.

    The question was quite simple “On scale of -100 to +100 how positive/negative do you feel about..? “

    The chart is based on the net figures for all those listed with a breakdown for “swing” voter – those who haven’t decided yet and those who say that they might change their minds.

      While Miliband continues to be a drag on his party the Tories are drag on Cameron and you can bet that both campaigns will try to build on their strengths.

    I do think there’s a tendency for the Tories to over-estimate the Miliband fact and to underestimate the ongoing appeal of Labour. Quite how this balance will work on May 7th will be decisive.

    The swing voters tab in the chart above has some quite good news for the LDs.

    Mike Smithson

    For 11 years viewing politics from OUTSIDE the Westminster bubble


    The sure guide to knowing if you live in a marginal – how much is coming through your letterbox

    February 17th, 2015

    The above polling by Ipsos at GE10 says a lot. By the final week just over a quarter of those sampled in its marginals polling realised that that was the status of their seat.

    I raise this because a prominent PBer emailed me to ask “What evidence is there that voters actually know that they are in a marginal constituency?” The answer is in the polling.

    My guess is that the term “marginal” is simply not understood by most voters. But does it matter?

    You know you are living in a marginal by the intensity of the campaign in your area. For those living in the 500 seats where the result is seen as a foregone conclusion will have a totally different expereince compared with those that do.

      Under first past the post winning seats is paramount not building up national vote share. The marginals matter – the rest don’t.

    In the constituency where I live, Bedford, the battle has been building up since the start of the year and over the past five days we’ve had four separate election communications. Heaven knows how much we’ll have got by 10pm on May 7th?

    How much of this will actually get through? How much will the huge financial advantage that the Tories have matter in key seats? Billboards are going to play a big part. We are going to see a lot on Twitter and Facebook and it will be interesting to observe how social media is used.

    But the one big election communication vehicle that is prohibited in the UK is paid-for TV advertising – totally unlike the US.

    Mike Smithson

    For 11 years viewing politics from OUTSIDE the Westminster bubble


    After six weeks of CON leads or them level-pegging today Lord Ashcroft poll has LAB 1% ahead

    February 16th, 2015

    So what do we make of all of this?

    After such a day of polling it is very hard to draw any conclusions. I must say that I was expecting Ashcroft to be in the same ball-park as ICM given they are both phone polls using very similar approaches with fieldwork over the same period.

    As well as the LAB lead it is the gap with ICM with UKIP that is very striking.

    I’m always reluctant to dismiss any poll as an outlier at such an early stage and we have not been able to study the ICM data yet. Ashcroft tables are already up.

    Labour will take a lot of heart from the Ashcroft positive/negative numbers and there’s little doubt that the party brand is much stronger that the leader one.

    In is interesting that UKIP/Farage are now at the bottom of that table. Maybe a pointer to anti-purple tactical voting?

    Mike Smithson

    For 11 years viewing politics from OUTSIDE the Westminster bubble


    UPDATED: Marf on the Copenhagen shootings and the day’s 1st 2 polls – ICM CON 4% ahead

    February 16th, 2015

    CON jump 6% in most sensational poll of 2015

    Still to come are Lord Ashcroft at 4pm and the usual YouGov.

    Last Monday both Ashcroft and YouGov had CON leads – since then it has all been LAB ahead.

    The LD 10% in the latest Populus has only been beaten by one other poll, ICM, this year.

    Mike Smithson


    Big question in blue-yellow battles is how much you can trust the Ashcroft 2 stage seat specific questioning

    February 16th, 2015

    Illuminating observations from the inventor – UKPR’s Anthony Wells

    There’s a big debate going on over the Lord Ashcroft style two stage questioning in his single constituency polls of which there have been more than 150. It will be recalled that after asking the standard voting question he puts a second one suggesting that those sampled focus on their own seat and the candidates who might stand.

    The numbers that are highlighted are the responses to the latter not the former. In his CON-LD seat polling the two stage approach is producing very striking results suggesting that there will be fewer easy pickings for the blue team than might appear from national polling.

    The issue, which could be crucial given the importance of the blue-yellow battleground to the outcome on May 7th, is how much the Ashcroft approach can be trusted. The good news is that it was tested in a limited fashion by UKPR’s Anthony Wells prior to GE10. He wrote recently:-

    “Personally I have confidence in the two-stage constituency question. It’s something I originally used in marginal polling for PoliticsHome back in 2008 and 2009, to address the problem that any polling of Lib Dem seats always seems to show a big jump for Labour and a collapse for the Lib Dems. This would look completely normal these days of course, but you used to find the same thing in polls when Labour were doing badly nationally and the Lib Dems well. My theory was that when people were asked about their voting intention they did not factor in any tactical decisions they might actually make – that is, if you were a Labour supporter in a LD-v-Con seat you might tell a pollster you’d vote Labour because they were the party you really supported, but actually vote Lib Dem as a tactical anti-Tory vote. The way that it only has a significant effect in Lib Dem seats has always given me some confidence it is working, and people aren’t just feeling obliged to give as different answer – the overwhelming majority of people answer the same to both questions.

    However the fact is the two-stage-constituency question is only theoretical – it hasn’t been well tested. Going back to it’s original use for the PoliticsHome marginal poll back in 2009, polling in Lib Dem seats using the normal question found vote shares of CON 41, LAB 17, LDEM 28. Using the locally prompted second question the figures became CON 37, LAB 12, LDEM 38. In really those seats ended up voting CON 39, LAB 9, LDEM 45. Clearly in that sense the prompted question gave a better steer to how well the Lib Dems were doing in their marginals… but the caveats are very heavy (it was 9 months before the election, so people could just have change their minds, and it’s only one data point anyway.) I trust the constituency prompted figures more, but that’s a personal opinion, the evidence isn’t there for us to be sure..”

    To my mind Anthony is right that the consistent pattern of this only showing big differences in LD seats suggests that something different is happening in those battles that should be taken into account.

    Certainly, looking at the betting prices in CON-facing LD seats, punters appear to putting their money behind the two stage approach.

    The LD are literally betting the party at the election on their individual candidates being the main selling point. They are effectively running “Mayoral election” type battles.

    Not long to wait till the early hours of Friday May 8th when we’ll know for sure what’s happened.

    Mike Smithson

    For 11 years viewing politics from OUTSIDE the Westminster bubble


    Labour are currently the largest recipients of tactical voting

    February 15th, 2015

    As part of this month’s Ipsos-Mori political monitor found that 22% of Lab voters saying they do so to keep another party out, rather than because the party represents their views. For the Tories the figure is 10%, for UKIP it is 9%and the Liberal Democrat’s it is 7%.

    Ipsos-More note “In previous years, the Liberal Democrats were most likely to say they chose their vote to keep another party out (16% of Liberal Democrats said so in May 2010, compared with eight percent each for the Conservatives and Labour).”

    In theory, this level of tactical voting should help Labour outperform the Tories in May, particularly with the unprecedented amount of constituency specific polling we’ve had in the parliament.

    On preferred coalition partners.


    The full data tables are available here.