So four, then.


A marginal figure which coincidentally matches the collective polling totals of Con, Lib Dem and UKIP.

Three or four or five.

But Labour had such a large swing/poll lead then things are different this time. They are polling not as far in front and the swings are not as big towards labour

The mandate for the next Scottish Referendum will go like this. SNP will take Scotland and the Tories will lose their last MP there next week. Then next year the remaining Tories including Ruth Davidson will also be wiped out. With no Conservative MPs or MSPs in place they can lawfully say Westminster under a Tory government have no right to their affairs as Scotland as a country did not elect them.

Yes then wins the Referendum all because of David Cameron and the Tories scare tactics. I would say Ed Balls, who it pains me to agree with, had this spot on, This is exactly how the SNP want it to play out. They are very manipulative and the Tories have fell straight into their trap.



People were offered AV but rejected it so this is the system they wanted.

Scottish Tories won't be wiped out - not a chance, Ruth Davidson is a very capable leader.

SNP at best will hold at 45% next year, but 2016 will be run on their record since 2011 - and the one thing they don't like is defending their own record.

They do need Cameron though to act as a smokescreen.

The last twelve opinion polls as a graph.

https://twitter.com/May2015NS/status/592722584107667457

So the polls don't really tell us anything apart from both party's are still neck and neck and neither have managed to pull ahead.

YouGov, Populus, PanelBase say Labour are winning.

Survation, ComRes, ICM, Ashcroft, Opinum say Conservatives are winning.

Looks more like a battle for which polling company is going to win, rather than which party.

What's the difference between their methods, which has been more reliable in the past, anyone know?

That will not happen. Under the Scottish Parliament system the Tories will always get seats.

Yes, under PR the Tories get seats based on their share of the vote don't they? So, even if they get 12%, they get 12% of the seats? Is that how it works?

It is a mixed system. Part FPTP, part PR. There is no way the Tories will get wiped out under that system. They will always get MSP's on the list system. (PR)

Quite right too. Even if the Tories are unpopular in Scotland, there will always be some Scots who have ideals that are similar to the Tories. It would be outrageous if they were never represented by anyone - ever. Especially if Independence does eventually come.

That is invalid because it cannot be verified. Also people change the minds very easily when the situation is right in front of them.

Polls that ask "Who will you vote for.." can be verified, by the result of the actual election.

But questions along the lines of "What do you think about this?" are so much hot air.

For example, I'm sure the "Bedroom Tax" had a 55% (or more support) just prior to it being implemented. It won't have that kind of support now will it?

When independence comes, I expect the Tory vote will rise in Scotland.

As far as I've been aware, ICM have generally been considered the 'gold-standard'. but looking at the polling in the last election it appears that many of the pollsters polled very similar results. In fact, it looks as if MORI may have polled the closest to the actual result:

This was the last ICM poll of the GE:

The actual GE result:

Lab: 29.7% Con: 36.9% LD: 23.6% Other: 9.8%

What pollsters last time out messed up was the LD share; all pollsters vastly over-estimated the support the LDs would get, polling them generally between 26%-28%, with the party actually losing 5 seats in the real GE, and polling at 23% - 7 percentage points below their highest polling mark (30%, in April with Angus Reid and a couple of other pollsters).

On ICM, I think the UKIP vote (and for that matter, the LD vote) could potentially slip them (and Ashcroft, too) up when polling. Apparently their recent poll shows many 2010 LDs moving to the Conservatives, more so than Labour which is a highly unusual sample. ICM post-GE 2010, got IndyRef and the EU elections wrong too - in the EU elections, they polled the Tories to come top, and Lab in third place as far as I recall - when the result was UKip, Lab, and Con. I think pollsters such as ICM and Ashcroft are underestimating the UKIP vote - I mean UKIP aren't getting 0 MPs at a GE - if worst comes to surprisingly the worst, Carswell will definitely hold Clacton - and other pollsters (well, mainly Survation tbh) are overestimating the UKIP vote. I sincerely doubt UKIP are going to poll 19% at the GE.

Survation and Ashcroft are relatively new kids on the block to polling, so this election will be the first time they're tried and tested. I also think Panelbase haven't polled at an a election before, either.

As far as I'm aware in this term (2010-15), ComRes have changed/adjusted their metholodgy twice (in fact in light of getting the LD vote spectacularly wrong, I think all pollsters may have made methodological changes in 2010 after the GE), YouGov have changed theirs fairly recently, in the last month or so, and ICM appear to have kept their general method - in which they reallocate based on previous voting history at the last GE (as does Ashcroft I think). I'm unsure about the rest.

That hasn't really been tested.


No kidding! According to that poll, Labour is now third in Scotland, behind the Conservatives.


I was actually referring to both of them. Sorry if that wasn't made clear!

The UK has a completely written constitution. It is in 25 Acts of Parliament and Documents.


http://en.wikipedia.org/wiki/Constit...ingdom#History

Technically you're both right: there are Acts of Parliament and other documents deemed "constitutional", but they are not codified into a single document that one would refer to as "the constitution" and do not have the protection that one would typically find for a constitution in another democratic country (for instance, the need for super-majorities and/or a plebiscite to change it). The Fixed-Term Parliaments Act 2011 is the first Act of Parliament (AFAIK) that requires a super-majority to make something happen; but that provision does not apply to the Act itself, which can still be amended or repealed upon a majority vote of both Houses, just like any other Act of Parliament.

For the course of the general election campaign, Newsnight each evening will be publishing an exclusive Newsnight Index on the likely outcome, based on a sophisticated forecast model.

It is produced by Professor Chris Hanretty from the University of East Anglia and his colleagues at electionforecast.co.uk .

Newsnight Index - 27 Apr 2015
Conservatives 281 (-5 from previous poll)
Labour 268 (+1)
SNP 49 (+1)
Liberal Democrats 26 (+2)
UKIP 2 (+1)
Green 1 (-)
Others 23

Previous poll:

18 NI Parties (5 SF, 0-1 Alliance, 9-10 DUP, 3 SDLP), 1 Respect, 3 Plaid Cymru, the Speaker? Or 2 PC and someone I've missed?

They give the breakdown for these three parties:
Democratic Unionist Party 8
Plaid Cymru 4
Social Democratic and Labour Party 3

Leaving a further 8 under 'Other' which I presume will be Sinn Fein, Alliance and Respect.

It seems very odd that the Speaker has to be elected to Parliament yet cannot represent their constituencies once they get there.

Anyone watch Panorama?

I was going to. I thought it would likely reveal nothing we didn't already know. So didn't bother.

Yes, this evening assorted polls of polls still put the two largest parties jointly on 33% or 34% so neither of them has been successful in breaking away.

Nat Silver has just announced his predictions:

Tory 283
Labour 270
SNP 48
Lib Dem 24
UKIP 1
Other 16

Given that UKIP will definitely win at least 2 it makes me worried about 'the worlds best pollster".

edit - typo on the ld score there.

Defeated.

Yes and the conclusion really is indicative of how close to call it is, I was hoping for something more conclusive but in a way was reassured it wasn't!