Richard Willis's Blog

First for local news and first for comment

New ComRes Poll – Tory Lead at 8%

There is a new poll from Communicate Research reported tonight for tomorrow’s Independent and ITV which shows the Lib Dems up a point and Labour down one as the Conservatives maintain a commanding 8% lead:

             Conservative             35% (no change)

                                       Lib Dem                    27% (up 1%) 

                                       Labour                      25% (down 1%)

This is very consistent with last night’s ComRes poll and is in fact statistically unchanged. I point out once again that ComRes have told me that these polls come with a  +/- 2% margin of error.

Commissioned by ITV News/The Independent, ComRes also asked voters who they thought would make a good Prime Minister. 45% said Nick Clegg, 44% David Cameron and 30% Gordon Brown. Asked if they had a good idea of what Nick Clegg stood for – only 56% said yes, while 65% said they knew what David Cameron stood for and 64% said the same of Gordon Brown.

Asked if they thought Nick Clegg would bring a fresh start for the country, 57% agreed, while 47% agreed when asked the same question about David Cameron, and 19% said the same of Gordon Brown.

ComRes telephoned 1015 GB adults between 19th and 20th April 2010. Data were weighted to be demographically representative of all GB adults. Data were also weighted by past vote recall.


April 21, 2010 - Posted by | Polls


  1. Wasn’t it ComRes who immediately after the first debate had a poll which had the Lib Dems on 35% in their raw data, but which was only 24% once they had ‘weighted’ it?

    I’d be interested to see their ‘weighting’ formula….. given all that has happened in the last week, I wonder if the variables in that formula are still as valid as they may have seemed when first developed.

    Comment by barcharter | April 22, 2010 | Reply

  2. You point out again incorrectly that the MofE is 2% . An O Level in simple Maths/Statistics will tell you that to get that level of MofE you need a sample size of circa 2,500 . Clearly comparing this poll with ALL the other polls produced recently fieldwork earlier in some , fieldwork at the same time as with Ipsos Mori and fieldwork later as with the Yougov/Sun , this poll is an outlier probably distorted by an ectreme sample of 500 on the 19th .

    Comment by Mark Senior | April 22, 2010 | Reply

    • I point out that ComRes have told me that your statement that the MoE is 4.5% is rubbish. They insist that their MoE is 2%. I will take what an impartial pollster says over a Lib Dem activist any day!

      Comment by Richard Willis | April 22, 2010 | Reply

      • I would be extremely surprised if you are correct in saying that Comres have told you that the MofE is 2% .
        They are members of the British Polling Council and would abide by their rules . On the BPC website there is a section ” A journalist’s guide to Opinion Polls ” , Point 8 covers MofE and completely refutes that claim .

        Comment by Mark Senior | April 22, 2010

      • Mark – I am sure you are not intending to accuse me of lying, although it does rather sound like it! Just so you understand what I am saying, when you started saying that the ComRes MoE was 4.5% I rang a friend who works at ComRes and asked about their methodology and the margin of error. I was passed to one of their technical people who told me that the MoE in the daily polls was 2%. I then asked about your statement that the actual MoE was 4.5% and was told that this was completely untrue.

        Now I suppose it is just about possible that the person I spoke to was mistaken but since I asked specific questions I think that unlikely!

        Comment by Richard Willis | April 22, 2010

  3. Anthony Wells in his report on this poll has the same conclusion , the sample on Monday the 19th had an unusually high Conservative figure , the samples from Sunday and Tuesday had figures more in line with the other pollsters but the Monday 500 sample has ended up distorting the results of 2 polls .

    Comment by Mark Senior | April 22, 2010 | Reply

  4. barcharter you clearly didn’t see the original release from ComRes last Friday which shows that the first results were from views and the second set were national. Looks like it was nothing to do with weighting at all

    Comment by dogbreath | April 22, 2010 | Reply

  5. Mark Senior if you are suggesting that Richard is lying then you are bang out of order and treading on dangerous ground. Unless you have some evidence to the contrary a good rule in life is to accept what someone tells you, especially if they put it in writing and publish it.

    Comment by Doodlebug | April 22, 2010 | Reply

    • Go to the British Polling Council and the relevant section on Margin Of Error calculations that I have pointed out , It is all there simply explained in black and white .

      Comment by Mark Senior | April 22, 2010 | Reply

  6. Interesting that Mark didnt point out last night’s YouGov poll which was:-

    Con 33 +2
    LD 31 -3
    Lab 27 +1

    This is in the same ballpark as the Com Res poll. But perhaps he is being selective with the polls he likes to quote?

    Comment by Doodlebug | April 22, 2010 | Reply

    • The Yougov figures you quote from Wednesday’s poll are exactly the same as in their Monday poll . You can average those 2 and the Tuesday one and it is clear that Comres over the same period is NOT in the same ball park .

      Comment by Mark Senior | April 22, 2010 | Reply

      • Mark – knowing how much you love the YouGov daily polls I am surprised you seem to have missed last night’s:

        Cons 34% (up 1%)
        Lab 29% (up 2%)
        Lib Dem 28% (down 3%)

        Further confirmation of the ComRes polls!

        Comment by Richard Willis | April 23, 2010

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: