« The EPP exit equations | Main | David Cameron on 7/7 »


Bromley put us at near rock bottom. It could be that previous polls were out of line, not this one.

Cameron's popularity is rising, but people don't associate his popularity with the Conservative Party.

The people who do vote Conservative in the past are feeling very wobbly about Cameron right now, as he has made so many left-leaning policy moves.

Could it be that we are losing traditonal support fast, and winning newer backers slow?

Why not move to a different strategy? Hold and build traditional support, while other parties disintegrate. We are more likely to win an election that way.


I'm not a professional statistician, but the larger the sample for your poll of polls the more accurate it is likely to be. If you exclude MORI then you will amplify the natural variation in the remaining polls you use.

I think you should retain the MORI poll as part of the POP. MORI may have very unusual results with a fair amount of fluctuation, but they are a major opinion poll and thus should be included.

Tim: so long as they are a mainline pollster, Mori should be in CH's PoP but their volatility is a serious problem. They were all over the shop in the election period too. It IS VERY INTERESTING that newspapers do not appear to be buying MORI polls at the moment.

I don't think the party should start changing strategy anytime things get challenging.

The party must continue to modernise and continue with DC's programme. There are three years before the next election so there is time. A lot of the high polling data lately is a reflection of the Government doing badly. There is nothing we can do about that. Unfortunately if the Government aren't doing very well then that means they are screwing something up which we are either paying for or using.

Feeble excuse to exclude inconvenient data? A bit like excluding Poll Tax from measures of Inflation. Very NuLab..

(Oh and by the way, William, I voted Conservative last time and feel not at all wobbly about Cameron, but then I'm a civilian not a party member.)

It would not be right to exclude MORI or any other major poll. Public opinion is obviously churning at the moment. The JP affair or affairs.The B&C result all could be having an effect. Let's be honest it's 3/4 years to the GE anyway, lots can change.

Don't strip it out, you're blinding yourself. The way to damp down a wobbly data source from wrecking your averages is to swamp it with others. Find some yet more poll sources to add in.

William, that sounds suspiciously like what we did under IDS - and look where that got us!

Cameron novelty is wearing off, press are getting bored, he is falling into the same traps as Howard (flash parties, trying to be cool) and has yet to deliver a single substantial or interesting position on anything. (the green stuff just looks bandwaggony and self indulgent, they dont give a f**k in Darlington).

I guess he will have a push in the Autumn, with no response, and then panic next spring before the locals.....

Where is the campaign for capitalism?
Where is the crusade on crime?
Where are the broad sunlit uplands?

I do find it quite odd that the right are already getting restless. I mean Cameron might be doing the whole change routine, but its not as if we've had a long string of left wing policies being published, we haven't had any real policies published. Personally I class my self as moderately centre, and I imagine Cameron will be keeping the right wing policies from the last election that were proven to hit home with the voters, such as limiting immigration. I wouldn't mind a few more policies being published though, especially the ones we know are certain to remain the same, like immigration, so that when someone asks me why they can join the Conservative party I have a better answer to give them than "Its not Labour".

"that sounds suspiciously like what we did under IDS - and look where that got us!"

And Hague, and Howard...

William said:
""Why not move to a different strategy? Hold and build traditional support, while other parties disintegrate. We are more likely to win an election that way."'
Eminently sensible. It seems that the MORI Poll indicates that the core vote is slipping away. If there's no delivery on the EPP pledge, support will weaken further. Tree hugging and the desire to be seen as 'nice' and unthreatening has a limited and diminishing appeal. Clear policies on the EU, immigration and taxation are essential to making progress in the polls. Too much effort is going into avoiding outbursts from Clarke, Heseltine, Gummer and the rest of the old Major administraton lags. So what if the BBC erupts in indignation? There's an election to win and sound policies are needed.

MORI is very volatile and isn't a poll I trust at all really. This is because its methodology is based on face to face random interviews on the street, which aren't weighted by past vote recall or social group as other polls are. Those surveyed are thus often highly unrepresentative. This benefits Labour normally, due to the random street sample and "shy tory" syndrome. It can however make it volatile as for instance, during a scandal planning on voting for the said party is not something many would admit.

The published figures are usually only those certain to vote as well (those people giving 10/10 chance of voting), whereas other polls include those giving lower probabilities and 'non voters' based on their voting history. Again, it makes it more valatile, especially as Project Cameron has reduced the chances of traditionalists voting (from 10/10 to 8 or 9 say).

I would happily see it taken out the PoP. That said it can identify sudden movements however YouGov is better at this I feel, particularly their new daily tracker surveys - hopefully covering voting soon.

The apparent flat-lining of tory support in the polls is nothing to be concerned about, and should absolutely not be used as a reason for returning to a core vote strategy.

We did poorly in Bromley for one simple reason - a very effective negative campaign by the LibDems. But their success will not be repeated at a national level: with Ming at the helm they are unlikely to break 20%.

We have a consistent (albeit slim) lead over Labour in the polls and should continue to feel positive. We've probably gone as far as we can on style alone and further improvements will come when we start adding substance.

David criticisms are correct - MORI is volatile because, firstly, it does not weight by past vote (or equivalent) and therefore has samples which vary in their political make up from poll to poll and, secondly, because their likelihood to vote filter is unduly harsh - entirely discounting people who say there are 9/10 certain to vote.

In fairness to MORI though I should point out that they do weight by other demographics, and they don't do interviews in the street - face-to-face surveys are carried out in people's own homes.

MORI should be retained but BPIX removed as they refuse to reveal details of their polls for analysis.

One thing I must do on BPIX, Jack W, is drop their last entry from the PoP if they don't report again soon. Their last survey was on 30th April and that (Con 35, Lab 32, LD19) is still weighing in the PoP.

I don't understand how Howard could be said to have followed a "core vote" strategy when the only issue he talked about at any length was immigration. Immigration is more a working-class concern than a middle-class one, and the Tories poll much less strongly in working class areas than middle-class ones.

Ergo, it's a fiction. The old Lenin line comes to mind about the lie told often enough becoming the truth...

Many years ago, I participated in a MORI street poll on political matters. Some way through the questioning, I glanced at the sheet the pollster was filling in, and saw that most of the boxes had been filled in completely incorrectly - I checked all the rest as I answered, and got the pollster to change the wrong answers.

I wonder how often that still happens?

Ed. I'd remove BPIX completely as they refuse to provide datasets of their polling and are not members of the professional body, the British Polling Council.

As for your PoP it would be better to take the last 4 or 5 polls as the measure rather than wait for each organization to re-poll. However there are better experts than I on Politicalbetting.com. Drop them a post and the experts will give you plentiful advice .... all of it free !

William, that sounds suspiciously like what we did under IDS - and look where that got us!

Posted by: Sally Roberts | July 07, 2006 at 16:39

It has taken some of the public and the media nearly ten years to fall out of love with the mostly incompetent and distructive Labour Government of Blair. Hague and IDS stood no chance; the media was still in love with Blair and even when Howard appeared the BBC were quick to remind us at his inaugeration that there was something of the blight about him. Cameron is just lucky with his timing - the media has fallen out of love with Blair - but not really gone overboard for Cameron and the further Left that he goes the more traditional support he (Cameron)will lose. What does Maude's "deeper and wider" mean? Maybe homosexuality will be made compulsory. Perhaps the Mods and the Progs will tell us.
IDS (a good man) made a mistake when he let May make a speech telling us all that when we previously voted Conservative we were voting for the Nasty Party and by inference we took it that we were not very nice: before I think of voting Tory again the Tories will have to return to its roots and become nasty again - in other words, effective. There are many ways that it can be done, but it will not be done if the Party moves to the Left, so why bother voting. It has been noted that there is hardly a fag paper between the mainstream Parties. What is Cameron going to do - move to the left of Blair and Brown? What does he want power for- an indexed linked pension?

John Hustings wrote:
""Immigration is more a working-class concern than a middle-class one, and the Tories poll much less strongly in working class areas than middle-class ones.""
What a contemptible attitude. So the working class, who bear the brunt of the moronic immigration policies, can just lump it as far as the Tory party is concerned. The Conservatives will campaign safely in the leafy middle-class constituencies and pretend that the immigration problem doesn't exist. There was a time when the Tory working class vote was significant. Once a national party, it now seems confined to middle England, and a sectional interest within its lager, at that. Occasionally, the stench of political decay on this blog becomes overwhelming.

Of course you should not drop Mori, so long as you weight for sample size. There may well be quite sizeable percentage errors, hence polls of polls. However, one cannot just make assumptions about reliability or otherwise - there are rogue polls.

MORI probably does need to re-examine its methodology, given last election, but it is still a reputable organisation and self critical, so keep it in but comment if its performance looks odd.

Worth occasionally comapring difference of individual polls from the mean.

"What a contemptible attitude. So the working class, who bear the brunt of the moronic immigration policies, can just lump it as far as the Tory party is concerned. "

No, that is not my view at all. I have consistently argued that the Tory Party should have a strong immigration policy. I myself have working class origins.

My point was to do with addressing the misconception that Michael Howard followed a "core vote" strategy at the last election. Since the Tory "core vote" is largely middle-class, this seems a totally wayward judgement.

That was the only point I was making.

Not sure this is worth getting too excited about. The polls show a 5% drop in a month. That doesn't mean a 15% drop in 3 months. These polls often have a margin of error of 3 percent. Keeping it in the 'poll of polls' is the best way of smoothing down these trends and reducing margin of error.

To go further on my last post, John Coles:

I believe that Michael Howard's policy at the last election was neither to target the core vote, nor to target swing voters/non Tories. I believe his strategy was simply one of emphasising Tory policy strengths. Since the Tories had a large lead on their immigration policy -- and on nothing else -- he chose to emphasise that policy, and avoid concentration on issues on which the Tories were less popular.

To some extent this is an understandable tactic given that he only had a short time as leader, and couldn't really spend too much time addressing public attitudes to Tory policies on health or education.

However, given the fresh start that David Cameron has had, as well as the ear of the public, I am very disappointed that he has gone down much the same route as Michael Howard and chosen barely to mention health, education or the economy except to allay fears.

So, I don't believe the problem of the last election was that there was a core vote strategy (there wasn't), but simply that there was an unbalance, and the Tories barely addressed the concerns of the public on a whole range of issues.

My hope is that the Tories restore some balance to their agenda in the future, and put some confidence and positivity into their strategy, rather than what we've seen up to now, which is a constant playing down of perceived fears.

Then I withdraw my observation about your first post, John. It seemed to propose an abandonment of the working class by the Conservative party and as someone with a working class background, that always makes my gorge rise. But I see that this was not the case.
My point on the need for policies (there still seem to be none) and valuing the core vote (instead of believing that it has nowhere else to go) stand: Mr Cameron and CCO have much to realise. There is more to getting the vote out than throwing Fund Raisers surround by pretty young creatures hired in for the occasion.
Excuse me whilst I put on my tie and the debate returns to whether MORI, the bearer of bad news, should be shot.

MORI should certainly be in the poll of polls, as they are leading polling organisation, belong to the British Polling Council, and (eventually) gave an accurate prediction of the General Election Result.

They are also the only pollster to ask people to ranks issues by importance, and consistently to poll attitudes to the party leaders.


"Maybe homosexuality will be made compulsory."

Thats it! And we're all going to be forced to sing the "Red Flag" when we see a red traffic light as well.

You certainly made me laugh.

Tim, I am a professional doctored statistician, and I have written often (and no doubt ad nauseam) about the lack of meaning (in terms of what can be inferred from) the "poll of polls". I am not going to rehearse the calculus again, but the number you publish is as valid an estimator of Tory support as would be gained by publishing the proportion of heads in 8 tosses of a fair coin plus or minus a random number from the interval [-1, .5]. Your concern about the visible between-poll variances isn't the issue - at least that variance is visible - what is concerning is that you have no idea about the hidden systematic biases. Many have written that this method is gonna be OK in the long run; they are implicitly appealing to something known as the central limit theorem, but it does NOT apply here. You should remove the poll of polls in order to maintain propriety for Conservative Home.

Never thought much of MORI.........Bob Worcester was too interested in selling out and had 3i along for the ride. Planning for sale means lots of Tv exposure and somehow it doesn't suggest accuracy or consistency to have that high profile approach - something about Medium and Message

Can I just say that the first sentance at least of William's posting (the very first one on this page) is complete garbage.

Firstly, by-elections are completely irrevelant when assessing the political landscape of the day, since turnout is often massively reduced and people know full well that they don't matter too much.

In Bromley the Lib Dems, who we already know have a very good by-election campaigning machine, managed to mobilise their election vote and actually increased it by 1,500 votes. Rather strangely, this coincided with an 8,000 plummet in the Labour vote - which meant that some Labour voters probably believed the Lib Dem (only the Lib Dems can beat the Conservatives here rubbish) and switched sides.

Of course, Ming the Moron and his minions made much of their failure to actually win the seat - since they know full well that they would have no chance of winning at a real election, and anyone who believes that the seat won't return whoever is the Conservative candidate at the next election by an enormous majority (as seen in 2005) is just plain dim.

Perhaps we should just not bother with opinion polls at all. They are quite often a means to swaying opinion rather than measuring it. It is very easy to say you will do something in a meaningless poll - but rather different when it comes to putting a cross in the box that will affect the future of the country.

Perhaps we should just not bother with opinion polls at all.

We should follow the dfrench system and ban them in the last 7 days of an election campaign

There are some concerns relating to the volatility of MORI's outputs.I do think however you need to retain them in the polling data.

Much of the data from all the pollsters tells me that the electorate have no strong allegianges.The recent bounce in Tory support being due to the incompetence of the Government.There is no sense that people are coming over to us to stay.

There is a general disconnect with politics amongst the electorate.The whole process is too remote and consistently failing to deliver real results.

This is what is so troubling about the DC project.I have no sense that he has anything to offer to fill this void.As such we will continue to get confusing and contradictory polling results from all Pollsters.This reflects a demoralised electorate who don't believe politicians and are not being offered anything positive from either side!!


You could always just weight the polls differently. Perhaps

30% YouGov
25% ICM
20% BPIX
15% Populus
10% Mori

That way you could incorporate all the available data and keep everyone updated as new polls come out, but Mori wouldn't drag the POP up and down so much with it's volatility.

The comments to this entry are closed.



ConHome on Twitter

    follow me on Twitter

    Conservative blogs

    Today's public spending saving

    New on other blogs

    • Receive our daily email
      Enter your details below:

    • Tracker 2
    • Extreme Tracker