Political polling

[redsquirrel]

Currently predicted to lose them.

Just by Mori or by others too? Correct me if I'm wrong but Mori are making that prediction on a Scottish UNS, has there been any constituency polling?

I'd be surprised if the Libscum did lose them, there would need to be a swing of about 25% from the LDs to SNP, and while their majorities were down the Libscum managed to hold onto both seats in 2011. And LDs even managed a plurality of the vote in EU 2014 elections, I know that's from before the referendum but I would still put money on the LDs holding them.

Orkney:

Britain First 37
British National Party 35
Conservative Party 679
Labour Party 466
Liberal Democrats 1825
NO2EU 14
Scottish Green Party 619
Scottish National Party (SNP) 872
UK Independence Party (UKIP) 618

Shetland:

BRITAIN FIRST 55
BRITISH NATIONAL PARTY 40
CONSERVATIVE PARTY 543
LABOUR PARTY 590
LIBERAL DEMOCRATS 1755
NO2EU 25
SCOTTISH GREEN PARTY 638
SCOTTISH NATIONAL PARTY (SNP) 897
UK INDEPENDENCE PARTY (UKIP) 616

 

[danny la rouge]

"The graph below shows the latest What Scotland Thinks/ScotCen Poll of Polls of voting intentions in Scotland for the 2015 UK general election and how it has evolved over time."
POSTED ON 29TH JANUARY 2015 BY JOHN CURTICE

http://blog.whatscotlandthinks.org/2015/01/poll-polls-westminster-vote-intentions-19-jan/

 

[Sue]

So the Murphy bump...?!

 

[William of Walworth]

In polling, excluding 'don't knows' is highly questionable in technical terms, danny la rouge ...

I'm not in any way denying that the SNP are storming ahead -- they are.

And I'd doubt at the moment that including 'don't knows' would change that too much -- if it did, it would probably not be beyond marginally so.

Just a note of caution is all though.

 

[brogdale]

So the Murphy bump...?!

As expected.

 

[Sue]

As expected.

Indeed though someone should've probably told the press that.

 

[brogdale]

In polling, excluding 'don't knows' is highly questionable in technical terms, danny la rouge ...

I'm not in any way denying that the SNP are storming ahead -- they are.

And I'd doubt at the moment that including 'don't knows' would change that too much -- if it did, it would probably not be beyond marginally so.

Just a note of caution is all though.

I heard Bob Worcester talk about this very issue at an LSE public lecture on Tuesday, and he told the story of how his, (then innovative) "leaners" analysis helped his client (Labour) in the Feb '74 GE. And, strangely, DK's are one of the methodological differences between the pollsters that YG's Anthony looks at in his excellent post on house polling differences...

Don’t knows. Another cause of the differences between companies is how they treat people who say don’t know. YouGov and Populus just ignore those people completely. MORI and ComRes ask those people “squeeze questions”, probing to see if they’ll say who they are most likely to vote for. ICM, Lord Ashcroft and Survation go further and make some estimates about those people based on their other answers, generally assuming that a proportion of people who say don’t know will actually end up voting for the party they did last time. How this approach impacts on voting intention numbers depends on the political circumstances at the time, it tends to help any party that has lost lots of support. When ICM first pioneered it in the 1990s it helped the Tories (and was known as the “shy Tory adjustment”), these days it helps the Lib Dems, and goes a long way to explain why ICM tend to show the highest level of support for the Lib Dems.

And these are just the obvious things, there will be lots of other subtle or unusual differences (ICM weight down people who didn’t vote last time, Survation ask people to imagine all parties are standing in the seat, ComRes have a harsher turnout filter for smaller parties in their online polls, etc, etc)

Well worth a read.

e2a :

It’s important to note that the pollsters in the middle of the graph are not necessarily more correct, these differences are relative to one another. We can’t tell what the deviations are from the “true” figure, as we don’t know what the “true” figure is.

 

[danny la rouge]

In polling, excluding 'don't knows' is highly questionable in technical terms, danny la rouge ...

I'm not in any way denying that the SNP are storming ahead -- they are.

And I'd doubt at the moment that including 'don't knows' would change that too much -- if it did, it would probably not be beyond marginally so.

Just a note of caution is all though.

I made no claims; I posted a new poll of polls and related blog post by John Curtice. The spread of companies used, the reasons for Scotland only polls being used, and other caveats, are discussed by Curtice.

 

[William of Walworth]

danny la rouge : Fair dos, but brogdale above (and Bob Worcester) confirms my more general point.

 

[danny la rouge]

You should re-read what Bob Worcester says, Will.

 

[William of Walworth]

danny la rouge : I've read most of the main Bob Worcester article now (from the link) as well as just the part about don't knows that brogdale quoted above. Not too sure what you're getting at?

 

[danny la rouge]

danny la rouge : Not too sure what you're getting at?

A). That is isn't Bob Worcester.
B). That it doesn't say it's "highly questionable in technical terms" to exclude Don't Knows. Indeed, the author (a YouGov employee) says that they prefer the YouGov way, of doing things, one of which is excluding Don't Knows. Indeed, the author says that trying to "squeeze questions" out of DKs "to see if they’ll say who they are most likely to vote for" has the effect that it "tends to help any party that has lost lots of support".

The conclusion, however, is "look at a broad average of the polls".

So, the article doesn't confirm your general point that "In polling, excluding 'don't knows' is highly questionable in technical terms".

(That said, I personally prefer to see the DKs included in the data. But I see nothing wrong with excluding them in order to see certain things more clearly).

 

[brogdale]

danny la rouge : I've read most of the main Bob Worcester article now (from the link) as well as just the part about don't knows that brogdale quoted above. Not too sure what you're getting at?

William of Walworth the linked article is by Anthony Wells of YouGov, and, amongst other things, describes the varying pollster methodologies for dealing with those polled who declare that they "Don't know" who they would vote for at the next GE.

The analysis that danny la rouge linked to is based upon a poll of polls, which to some extent, negates any particular issue of individual pollster DK methodology; that's one key advantage of employing a poll of polls approach.

I suspect that's what he's getting at; your criticism appears mis-placed.

 

[danny la rouge]

Exactly so, brogdale.

Also, it appears that yesterday was #SarcasmThursday chez la rouge. Sorry William of Walworth.

 

[brogdale]

Exactly so, brogdale.

Also, it appears that yesterday was #SarcasmThursday chez la rouge. Sorry William of Walworth.

FWIW listening to Worcester's account of his work for Wilson in '74 was (to me) fascinating. He had many of the old, original OHP transparencies and ring-bound folders of data that he personally presented to Wilson, Healy and the Union leaders involved in the election strategy meetings. Worcester said that Mori's development of pushing DKs into how they 'leaned' helped Labour tailor their messages in the marginals.

A couple of other points he made were also quite revealing; one finding that was fairly high on the list of respondents priorities for the LP was that they were "moderate". I've long suspected that the retreat to centrism would have been reinforced by poll findings, and Worcester explicitly showed that pressure, even in '74. Another Worcester anecdote that struck me was the one when he had to inform the Dep.Gen.Sec. of the LP of the extent of 'middle-class' support for the party; the response was (apparently)..."we don't need middle-class support, we're a party of the working class!". Hmmm

 

[brogdale]

Should also add that, along with Bogdanor, the psephologist David Butler was on the panel. He did speak for about 10 mins during which he recalled the 1931 general election; at which time he was 7 years old!

 

[danny la rouge]

Who was Dep Gen Secy that time? Bill Simpson?

 

[brogdale]

Who was Dep Gen Secy that time? Bill Simpson?

Can't remember who Worcester said it was. Sorry.

 

[Mr Moose]

Since this thread started the Labour lead has narrowed alarmingly. Ed, I'm sure, will be better in debate than portrayed in the press, but Labour need a strong campaign now.

 

[Dogsauce]

Doesn't help with the Blairites knocking chunks out of the party's campaign on the NHS. It's almost as if the pricks want to lose so their worldview can be proved right, that Blairite 'centrism' is the only way to victory. There is no alternative.

 

[Mr Moose]

I'm sure there is truth in that, but the Tories are also rallying.

 

[danny la rouge]

I'm sure there is truth in that, but the Tories are also rallying.

Not enough thus far.

 

[William of Walworth]

Apologies to danny la rouge and brogdale both. I was an idiot this morning -- tried to rush reading and posting in much too short a time.

I still do think there are issues with excluding Don't Knows though. Will reread and then get back to this another time with a better brain on. Ahem, etc.

(Due at railway station in just over half an hour. A bit like my bus to catch this morning, leaving time of it was 50 minutes off and further to walk. So unlike this morning, I'm not going to even start to attempt anything now )

 

[killer b]

Forgive me if I've got the wrong impression William, but there seems to be a problem in your eyes for any method that doesn't produce a Labour lead.

Do you think there should be a wishful thinking factor applied?

 

[William of Walworth]

No. I'm fully aware they're doing badly atm and haven't denied it. You have my thinking entirely wrong,

Perhaps the above is your wishful thinking about what you want me to be thinking? <-- about as accurate as your speculation probably

 

[Mr Moose]

Not enough thus far.

It's not a good trend on the UK polling site graph.

 

[killer b]

Wishful thinking about wishful thinking? Hey, dawg...

 

[danny la rouge]

Wishful thinking about wishful thinking? Hey, dawg...

This is getting too recursive for me!

 

[William of Walworth]

No wishful thinking here, except about wanting the Tories not to win -- and hopefully do more badly than the polls say. That's another matter altogether. Would also like the LibDems to be crushed electorally.

We'd probably agree a lot more in our antipathy/objections to Labour than you think, but let's leave just a bit of space for specific Tory-hatred.

Right, got to dash!

 

[kabbes]

The problem is that hoping that the Tories and LibDems do badly necessarily entails hoping somebody else does well.