Many years ago Teletext (those of you over 25 will remember it) used to have phone in polls on issues of the day. On occassion they would ask voting intention, and it would invariably show the Conservatives on about 80% of the vote even in the midst of Blair’s greatest popularity – presumably because only elderly Tory voters bothered to ring into Teletext polls.

I was rather reminded of it by this from Sky News. Conducted on their own panel it has voting intentions of Conservative 43%, Labour 24%, Liberal Democrats 8% – repercentaged to exclude don’t knows and wouldn’t votes, it works out at CON 50%, LAB 28%, LDEM 10%, Others 13% – so while reputable pollsters are showing a Conservative lead of between 2 and 6 points, Sky’s panel are showing a lead of 22 points. That rings alarm bells to say the least.

This isn’t actually a voodoo poll in the purest sense, it was conducted using a panel, rather than an open access “red button” poll (although there is no indication of whether there was an attempt to draw a representative sample from within the wider panel) – but the sample looks very ropey and there is no apparent attempt at proper political weighting. There are sparse demographic details in the results, but the 2010 recalled vote break shows 44% of the sample voted Tory, compared to 17% Labour and 18% Lib Dem. For context, established polling companies like ICM weight their polls so that 25% of the sample is people who voted Tory in 2010, 21% Labour and 16% Lib Dem.

You sometimes get fun little red button polls on media websites, but they normally come with disclaimers that they are not properly represenative polls. In contrast, Sky have it as the headline on their website, liberally sprinkled with quotes from their Chief Political Correspondent Jon Craig about what it would mean if repeated at a general election. Sigh.

Ignore (and for journalists out there, this summary by Peter Kellner from the BPC website about when to pay attention to a poll is always worth revisiting).

UPDATE: Jon Craig’s blog here at least starts by acknowledging “Now I know the sniffy ones among you – yes, you know who you are – will say it’s not a wholly scientific, weighted opinion poll and all that.” On one hand, I’m pleased he’s added the caveat. On the other hand, one is rather tempted to reply that you shouldn’t bloody publish it then. Wanting polls to be scientifically weighted is not some odd personal fetish or the pedantry of pollsters and statisticians in ivory towers, it’s that all that makes a poll meaningful is that it is representative of the wider population, through proper weighting and/or sampling. A poll that doesn’t do that is just the views of an arbitary 1500 people, who do not necessary represent anyone but themselves.


74 Responses to “No, the Liberal Democrats are NOT on 8%”

1 2
  1. @ROB – ICM Guardian poll for 9th August 1997- same moment in political cycle:
    Lab 55
    Con 29
    LD 12
    OTH 4

    Nothing to add really – I just wanted to post it again. ;)
    Best not get too happy about the state of the polls though lest steam starts blowing out of Roland’s ears.

  2. Julian – That was funny.

  3. Eoin,

    Iraq dead as a political issue? I doubt it when it appears that there is likely to be an Inquest on David Kelly’s death.which will give a lot of publicity.

    And Tony Blair’s book launch.

    I suspect that in private Tony Blair is decent enough to know he did the wrong thing; but the invasion of Iraq is not something one can say sorry about.

    As I have blogged on this site before,a live psephological issue is how Labour can convince the electors that their next and future leaders have sufficient courage to stand up to the likes of George W Bush when necessary, and that will take some doing given their government’s record on Iraq and Afghanistan.

  4. One poll doth not a summer make.

    Look at the situation one year ago, Labour nowhere. They could have almost won in May had they followed AD’s advice (but he should have given a Budget sweetener to Mum’s – difficult for Scots chauvinists, among whom I do not class GB, bless him).)

  5. Frederic,

    In terms of barring yellows returning ‘home’ to reds is how I meant it. I quite imagine for Iraqis it will never be a dead issue.

    Also, like any issue on the index of importance to voters it is all relative. Did you know that Iraq was not even on the top ten issues in May 2010? I imagine it is even further down the list now.

    If this was a counselling forum i would not have termed it as such. I meant dead in the psepological sense.

  6. @Andy JS

    You said “…this is an embarrassing mistake. I can’t believe they’ve actually made it – I can understand it getting so far in their editorial process but to actually go all the way through and out the other side is astounding…”

    Your faith in Sky’s devotion to truthful reportage is misplaced, Andy: it’s no mistake, it’s deliberate. Ever since the LIBs played LAB against CON in the Coalition negotiations, the press have been working on tearing Clegg a brand new wide one. See Roy Greenslade’s articles at the time:

    ht tp://www.thisislondon.co.uk/markets/article-23833069-tory-press-will-not-make-life-easy-for-nick-clegg.do

    Regards, Martyn

  7. @ROB SHEFFIELD

    You said “…2013 is my prediction for the creaky edifice to collapse…”

    I thought I’d put this one to bed in previous posts. In the UK, coalitions last the full term, minority governments collapse prematurely. Don’t make me go back and recite the whole list again, my head’ll explode.

    Regards, Martyn

  8. @Bobby

    You said “…ICM Poll: Tories – 37%, Labour – 37%, Lib Dems – 18%…”

    Even with my “snog YouGov, marry ICM/Populus, avoid ComRes” head on, that’s too high for the LIBs. Unless there’s a whole new breed of shy LIBs, (well, shyer… :-) ) evolving, and we won’t know that until some elections pop up. What were they in the last ICM poll?

    Regards, Martyn

  9. @virgilio

    You said “…Go Julia!!…”

    At the risk of thread-interleaving, those genuinely curious about AV may wish to examine the upcoming Australian elections, where the Welsh girl plays against the guy with too-tight Speedos and too much belly hair (ew…)

    Regards, Martyn

  10. @Billy

    You said “…What’s the actual maths (or science) behind the weighting of polls, like the calculations and stuff…”

    If you take a bunch of people (your “sample”) and ask them questions, you’ll get a fair idea of what they think. But you don’t care about what they think, you want to know what everybody (your “population”) thinks. There’s some heavy duty maths (the “Central Limit Theorem”) that says you can get from one to the other, but for that to work, your sample must be roughly similar to your population (a “representative sample”).

    But life is awkward and inevitably, your sample will be different from your population: there’ll be far too many teenage males with PCs at Uni and far too few fiftysomething women, for example. Problem is, fiftysomething women vote like their life depended on it and teenage males wouldn’t vote if you pointed a gun at their gentlemanly bits. So this is a major problem.

    So you fiddle your sample a bit (“weight your sample”) until it looks roughly like your population, then fiddle with it some more (“weight for propensity to vote, recall bias, etc”) to round off the edges. You then slap the results on a spreadsheet, contact the media, and go to the pub. Result!

    Sky didn’t do the compensatory fiddling and as a result, the poll smells of poo. Jon Craig isn’t paid to tell the truth, he’s paid to make Rupert smile, so he isn’t bothered. You, however, presumably want to be told the truth at some point in your life. Which is why you’re here and Jon is there.

    Regards, Martyn

  11. Martyn,

    Yellows were 19% in the last poll (down 1%)
    Saying that ICm have reds +3% and blue -1%

    A very crude transferance onto YGs trends (Sorry Jay ;) ) would leave use with a 40% / 39% 15% – (others much to low even for YG/ICM in my view)

    Let’s wait and see it is out at 10pm.

  12. Howard,

    I fully realise that one poll isn’t a game-changer in the slightest.

    I also support Leeds United, and like supporting Labour, in recent years you need to get what joy you can when you can!

    The next step is a good solid poll lead and the acceptance of the New Leader. A good showing in the next elections is the crucial.

    Finally, a new policy base is required, and at least a year for them to be accepted by the public as the policies of a Government in waiting.

    Easy ;-)

  13. @MARTYN -“In the UK, coalitions last the full term…”
    @ROLAND – “This amazingly popular coalition will, in my opinion probably do a full term.”
    I agree. I think they will last the full term too.
    Exactly ONE full term.

  14. “In the UK, coalitions last the full term”

    With such frequent and recent examples to go by Martyn, your certainty is unshakeable.

  15. Does anyone watch Sky news for actual news these days? I thought it ceased to pretend to be a serious newscaster ages ago when it employed the likes of Craig, Burley, Boulton and co.

  16. @ Martyn

    Thanks for that – I’ll give the Central Limit Theorem a look.

    My actual interest (which I didn’t make that clear) is in the weighting – how they calculate the weightings for each group (I imagine gender, age, finances etc. must play a part, but translating that into percentage points here and there).

  17. Graeme,

    I aint seen Sky News since July ’97

  18. In the Ipsos poll someone linked, would the question on spending cuts be considered a leading question? Just interested to know.

  19. @Sue Marsh

    [Regarding my contention that UK coalitions last], you said “…With such frequent and recent examples to go by Martyn, your certainty is unshakeable…”

    Thank you. :-)

    However, treating your point seriously, the door swings both ways. We don’t have recent UK examples of coalitions lasting, but we don’t have recent UK examples of coalitions collapsing either. Of the only UK examples we do have (mid 19th century to 1945) they last pretty well.

    Regards, Martyn

  20. Yet

  21. ANTHONY [email protected]

    “repercentaged to exclude don’t knows and wouldn’t votes, it works out at CON 50%, LAB 28%, LDEM 10%, Others 18%”

    Strange – this comes to 106%!

  22. BT says –

    You’re right. The others should be 13%, not 18%.

  23. :cry

  24. :cry:

1 2