UK Opinion Polls Thread

[YL]

Here are the percentages from the polls conducted in Sheffield Hallam and Eastleigh:

Sheffield Hallam:

Lib Dems - 33%
Labour - 31%
Tories - 28%

Eastleigh:

Tories - 42%
Lib Dems - 31%
Labour - 21%

http://ukpollingreport.co.uk/blog/archives/2834/comment-page-1#comment-680315

Constituency polls in this country are, of course, known for their outstanding accuracy.

LOL. This is just too good to be true. I hope it's true.

As someone who lives there, I think the Hallam one is quite plausible, actually.  There's a lot of annoyance at Clegg, and Labour made the right choice of leader to take advantage.  (That isn't saying that they made the right choice of leader for other places.)  Indeed, the fieldwork for this was done before the Browne report (although probably after the recommendations started being leaked) and if the poll was accurate, I suspect Labour are now ahead in Hallam.  The council elections next year could provide some indication of what's going on; presumably there will be some attempts by the local Lib Dems to distance themselves from the Government.

As far as I'm aware, there really haven't been enough constituency polls in recent UK general elections to judge their accuracy record.  Is there a list anywhere of results?

To add a note of caution, shouldn't this be treated with some of the scepticism due a Tory internal poll, given the source?

[YL]

There was also a poll conducted and released in April this year in Norwich South, which showed Charles Clarke with a big lead, and the Lib Dems tied with the Greens for third place....

I'd forgotten that one: certainly not a triumph.

There is a document on the British Polling Council website
http://www.britishpollingcouncil.org/simon-atkinson.pdf
which gives various polling statistics from 2005 and includes 5 constituency polls, three by ICM and two by NOP.  To sum up, the ICM polls (in Finchley & Golders Green, Shipley and Haltemprice & Howden) were pretty accurate (they got the Shipley winner wrong, but it was close enough that it can't be considered a bad poll) while NOP were accurate in Cardiff North but badly overestimated the Tories and underestimated an Independent (and Labour, to a lesser extent) in Ynys Môn.

[YL]

There is a redistricting prior to the next elections, right? How will it affect the current bias towards Labour?

Off topic for this thread, but:

No-one really knows at the moment.  It'll probably hurt Labour a bit, partly because Wales, where Labour are strong, is currently over-represented and this is being ended, and partly because some more Labour-inclined areas of England are losing proportionately more seats.  (NB the idea of over-represented Labour inner cities and under-represented Tory suburbs and countryside is over-simplified: many inner city constituencies have been growing in the last few years, and e.g. Manchester Central now has one of the biggest electorates in the country.)  However, my understanding is that most of the "bias" to Labour is a differential turnout effect, not one due to the constituency sizes, so it probably won't change things in the Tories' favour as much as some of them seem to think.

[YL]

http://electoralcalculus.co.uk/userpoll.html

Electoral Calculus have changed their UNS calculator to include the new boundaries.

You might enjoy what happens to Sheffield West & Penistone.

[YL]

As Hallam exists now, Clegg'd just hold on (probably), but it'd be hard to tell how much the 2010-Cleggasm has receeded there (meaning, it could be worse or not as bad as the rest of the nation). The problem for Clegg with Sheffield though is that it's a mix of the exact group who've swung against the Liberals the hardest - students (there's two massive universities within a 5 minute drive of eachother). Come 2015, there'll be no student in Sheffield who hasn't seen that photo.

And if the new boundaries go through Sheffield West & Penistone, as it'd become, would be much less friendly to Nick Clegg.

I think you're basically right: as things stand Clegg would just hold on in the seat as it is now but would probably lose (to Labour, who are helped by the removal of Dore) on the provisional new boundaries.  (The seat which Dore goes to, basically a heavily redrawn Sheffield Heeley, might be quite interesting too: it's notionally Lib Dem with a smaller majority than West & Penistone, but the Lib Dem vote held up better in parts of Heeley last year than it did in Hallam.  I presume Clegg would stay with the bulk of his old seat though.)   But 2015 is a long way off, and if there's an election before then the events leading to it might affect his chances...

(Plus there's a small chance that we might get more sensible boundaries, in which case the calculations would of course be different.)

[YL]

Presumably a bit of an outlier, but tonight's YouGov has Labour on 45%, 13 points ahead of the Tories.  Lib Dem retention is down to 29% (there's going to be a large margin of error on that, though).

http://cdn.yougov.com/cumulus_uploads/document/21o4h4zdaz/YG-Archives-Pol-Sun-results-190412.pdf

[YL]

Instead of asking people how they were going to vote, ICM (for the Sunday Torygraph) asked them what they thought the result of an election would be.  There were two questions: first they asked "as a bit of fun" without giving the 2010 result, and then they gave the 2010 result and asked again.  Without the 2010 result, the mean answer was Lab 40, Con 32, LD 14; with it it was Lab 39, Con 31, LD 16.

Apparently they're going to publish a paper on this in the International Journal of Market Research next month.

http://www.icmresearch.com/wp-content/blogs.dir/1/files/2012/06/Wisdom-website-June-12.pdf