(I’m on holiday in Dorset at the moment and this guest slot submitted by Flockers echoes quite a lot of comments that we are getting whenever YouGov polls are published - which is now five times a week. When I return next week I’m hoping that the firm’s Peter Kellner will take part in an online Q&A session - no doubt this will form part of the conversation. I should emphasise that these are the personal views of Flockers - Mike Smithson)

What is it about their weightings?

For good reason, all polling firms apply weightings to their raw data in order to derive the final figures. There is little value in seeking to predict the responses of the electorate from the responses of a sample if that sample is not representative of the electorate as a whole; a poll taken in the Birkenhead working men’s club would have very little predictive value.

Looking at the YouGov data, I was immediately struck (see table above) by the resilience of the unweighted Conservative vote share and of the unweighted Conservative lead over Labour.

In each of the eleven 2010 polls to date, the unweighted Conservative share was in the range 41-46 (reported 38-40) and the Labour share in the range 26-30 (reported 30-33). The average unweighted Conservative lead was 14.4 points against an average reported lead of 8 points, indicating an average weighting of 6.4 points away from the Conservatives.

Crikey! If YouGov are habitually over-sampling Tory supporters, how massive must the unweighted Conservative lead have been when Labour support slumped in the aftermath of the expenses scandal?

Not so massive. In seven reported polls during May and June the unweighted Conservative share was in the range 38-44 (reported 37-43) and the Labour share in the range 21-26 (reported 21-27).

The average unweighted Conservative lead was 15.7 points against an average reported lead of 15.7 points – the weighting against the Conservatives was 0.012 points. The weighting moved in the Conservatives’ favour in three polls, against them in four.

When did YouGov’s weightings start causing the weighted result to diverge so dramatically from the unweighted result? Quite recently:

In 16 reported polls for which the fieldwork was completed in July, August or September the average unweighted Conservative lead was 14.4 points against an average reported lead of 12.8 points, indicating an average weighting of 1.6 points away from the Conservatives, far below the 2010 average weighting of 6.4 points.

The weighting moved in the Conservatives’ favour in four polls, against them in twelve. During October, November and early December, 15 polls gave the Conservatives an average unweighted lead of 15.5 points, against an average reported lead of 12.6 points, indicating an average weighting of 2.9 points away from the Conservatives.

In the twelve subsequent polls the average weighting against the Conservatives has been 6.3 points.

Every single poll in 2010 has seen the Tory share weighted downwards and the Labour share weighted upwards, with a net effect on each poll of between three and nine points. Ten of the eleven polls this year have seen weighting impact the Tory share by more than four points. Only five of the previous 39 polls was adjusted by the same amount.

The previously close correlation between the score produced by an analysis of the unweighted responses and the weighted results has been dislocated.

Why? If the weighting’s haven’t changed (and I understand that’s what YouGov will say), then the only answer can be sample bias. Have all eleven of YouGov’s 2010 samples been heavy with people having Conservative-leaning characteristics (but who, oddly, are not influencing the numbers saying they will vote Conservative)? YouGov does not weight by certainty to vote – which would be one credible explanation if this discrepancy was seen in another pollster’s work.

A historical perspective But surely none of this matters, because unweighted samples are irrelevant. Well, quite.

But what if I were to tell you that in six of YouGov’s last seven polls before the 2005 general election a simple analysis of the unweighted sample would bring you to within a 2% Lab overstatement of the actual result? Not bad for an unrepresentative sample. As it happened, the 2005 polls saw on average a two point weighting to the Conservatives, which made YouGov one of the most impressive performers in the 2005 general election.

By way of further historical perspective, a random sample of eight 2008 polls produced a mean weighting against the Conservatives of 2.3 points, and a random sample of five late 2007 polls produced a mean weighting against the Conservatives of 2.6 points. The current level of weighting appears to be unprecedented for YouGov in the Brown/Cameron era.

Good news for the blues? Not entirely. The unweighted samples support the recent narrowing of the polls, with the Conservative lead in the last three polls averaging 12.33 per cent, against an average 17.9 per cent lead in the first three polls of the year. Indeed the reported numbers may even have masked the extent of the narrowing, as the polls early in January were some of the most heavily weighted in the sample. The data also shows a steady increase in Labour support since May.

However, if you are prepared to look through the weighting, the basic unweighted responses of people polled by YouGov still point to a commanding Conservative lead.

Flockers

Could this be where power ends up resting?

(Anybody who watched last week’s Question Time from Belfast would have been left in no doubt about the complexities and continuing tensions within Northern Ireland politics. With all sorts of different developments that could have an impact on the UK outcome I’ve asked long-standing PBer, Yokel, to give us a run-down. In this first part he looks at the big picture - then in future posts he’ll examine the eight critical seats. Mike Smithson)

It’s a thought to make the chattering classes in GB shudder. Could the political parties in Northern Ireland really hold the balance of power after the next GE? The Unionists will likely back the Tories in the event of tight parliamentary arithmetic given that Labour hasn’t done them too many favours. That won’t stop Labour trying though, despite the SDLP links, if the feelers put out to Lady Sylvia Hermon in North Down are anything to go by.

Until recently the eighteen Westminster seats were pretty easy to work out. This time things are more fluid, particularly on the Unionist side. Despite the DUP taking hit after hit, to profit from another’s disarray you need someone to take advantage.

The Conservative/Unionist alliance (hereby known as the CU’s) have not made the most of the opportunity thus far and has its own problems. Within Nationalism it’s all about whether the SDLP can stem the Sinn Fein tide, hold South Down and prevent any Sinn Fein growth killing off Alistair McDonnell in South Belfast. The great unknown is whether Sinn Fein will suffer stay at homes with its own issues.

The DUP does face a possible squeeze. At one end is Jim Allister’s TUV who are only going to stand in some safe unionist seats (Allister won’t confirm where) and the TUV has the weakness of being somewhat a one-man party. (Jim is on the far left in the QT screen shot above).

A recent poll put it on 6% compared to the DUP 18%, CU 15% and PUP 3%. NI Polls are notoriously unreliable but the basic direction of traffic says the DUP have lost a lot of votes, yet the CU’s are not really gaining, at least not across the board.

With TUV support fairly strongly concentrated, both they and the PUP wont be standing everywhere so it could cause the DUP a headache if the CU’s gain votes in the right places. Then there is another possibility. Could a weakened DUP, a stronger CU and stay at home Unionists (a big danger) let Nationalists slip into some seats?

There are many unknowns, the current political process has twists and turns to come in the next two months and it is unclear exactly who the CU’s are putting up as their PPC’s. Much, therefore, could affect things between now and election time.

The current Westminster position is DUP 9: UUP 1: SDLP 3: SF 5. The spread on DUP seats on the Betfair party line market is currently 6.5 - 8.5 seats.

I will cover eight potential battleground seats in Part 2.

Yokel

****Great Antifrank piece on the seat markets on PB2****

The map that changed my view of the election

I’ve shown this before but it’s worth running again as we try focus on why things might be different in the marginals.

For combined with the Andy Cooke analysis and the marginals polling, including last week’s aggregated MORI data, a clearer picture is emerging of the dynamics of the coming election and a better sense of the outcome.

The map first appeared here in September 2007 in a guest slot by geographer, Blair Freebairn, in which he tried to show where the election would be won and lost. This had a great influence on my thinking which has been reinforced by what we’ve seen in the past few days.

Blair noted “Take a good look at the map. Notice anything? That these marginal seats will decide the next election is not news. But look at the pattern the 201 marginal seats highlighted make. They don’t concentrate in Wales, Scotland, London, the major cities or the truly rural areas. They aren’t really regional. They are heavily concentrated in Medium English Towns and Their Hinterlands (METTHs from now on).

From Scarborough via Stourbridge to Hereford. Or maybe Cleethorpes to Halifax. Stevenage to Swindon by way of Luton. From St Austell to Taunton and up to Stroud and Redditch. Kettering Corby and Broxtowe (hi Nick). How about Gravesham, Hastings and Basingstoke. Burton across to Southport via Chester. The marginals are strung like bunting through Britain avoiding the cities and the truly rural. It’s the towns, stupid!

These seats are clustered on a fine scale but not a large one, in other words they occur across all parts of the UK but where they do occur you get lots of them….”

Last month Blair added further to his insight when he observed while the Cameron family debate was in full flow that significantly more couples got married in the marginals that Labour if defending than in the party’s strongholds.

So we start to get a picture of what a key election battle-ground looks like. It’s a medium sized English town, which is not part of a big conurbation, and was generally held by the Tories until the great Blair landslide of 1997. By contrast with the cities it is more socially cohesive with a distinct demographic profile.

Politically they are likely to have seen a big decline in the number of Labour councillors and activists in the past decade or so.

It’s how the party messages play out in places like this that matter most - for this election is decided not in the 650 seats but the 100 marginals starting with target number 51 on the Tory list.

So policies designed to help the cities and the concerns of the big conurbations matter less in electoral terms than designed for the towns -

Hopefully we’ll have a marginals poll from Angus Reid in the next couple of weeks.

Mike Smithson