The daily YouGov poll for the Sun tonight has topline figures of CON 36%, LAB 38%, LDEM 8%, UKIP 11%. The two point Labour lead is the lowest that YouGov have shown since October, and 36% is the highest they’ve shown since October. Usual caveats of course apply – it could be a further narrowing of Labour’s lead, or could just be normal margin of error. It does, however, underline the narrowing of the Labour lead that we saw in YouGov’s daily polling last week.

Meanwhile, as if to illustrate how much of the daily back and forth of polls is just random variation, this morning’s twice-weekly poll from Populus shows movement in the other direction. Topline figures there are back to CON 33%, LAB 40%, LDEM 13%, UKIP 8%. Tabs are here.

Finally there is another batch of UKIP donor Alan Bown’s Survation constituency polls, showing high levels of UKIP support in most of the seats selected – even compared to Survation’s national polls, which tend to show the highest levels of UKIP support to begin with. All four constituencies surveyed have much higher Con=>Lab swings than national polls imply, to a extent that looks somewhat doubtful to me. Swings at general elections aren’t uniform… but it’s a fair guide, parties perform a little better in one seat, a little worse in another seat, but if you’ve got a series of polls showing swings that are *all* substantially better than the national average, almost regardless of marginality, who holds the seat, etc, something’s not right. Somewhere or other they need to average out.

These seats where presumably selected as ones where they thought UKIP were doing particularly well, so perhaps that’s the reason – where UKIP are doing particularly well it results in a bigger swing (in which case they would by definition not be typical of other seats – so do be careful of extrapolation) but I’m dubious about constituency polling so far from the national picture, especially without political weighting. We shall see.

The most interesting thing I actually found there was the difference between the increase in the UKIP vote in the three coastal towns polled (up 23, 20 and 25 points) and in Crewe and Nantwich where it was up only 8. Now, leaving aside the prompting and the weighting and whether it’s a good measure of the actual level of UKIP support, all four were done on the same basis so should be comparable to each other. One interesting question about UKIP support at the next election is how uniform it will be – UKIP got comparative few council seats in 2013 for the level of support they achieved because it was spread so evenly, they just ended up coming second a lot. If their support in 2015 is the same they would struggle to translate support into any actual MPs. In terms of winning seats it’s much better to have areas of strength and weakness. Seaside towns were some of their better areas in the 2013 locals, and the contrast here between Crewe & Nantwich and the seaside towns suggests their support may be clumpier than thought… but again, don’t read too many conclusions into that single poll.

This morning’s YouGov/Sun poll has topline voting intention figures of CON 32%, LAB 39%, LDEM 10%, UKIP 12%. It also asked how people across Britain as a whole would vote if they could vote in the Scottish referendum – 22% would vote for Scottish independence, 55% would vote against, so more opposed to independence than Scotland itself (obviously the poll included a Scottish cross-break, but I’d caution against reading too much into that – stick to proper, bespoke Scottish polls for that, I suspect there will be plenty along in the aftermath of the white paper). Full tabs for the YouGov poll are here.

There is also a new Survation poll of Thanet South (tabs here), the first of a series of constituency polls commissioned by Alan Bown, a major UKIP donor, presumably of seats they see at potentially good for UKIP. The rest are likely to come out in December, but this one is out early because of Laura Sandys announcement that she’s too retire (though the poll itself was mostly done before that).

Topline voting intention figures in the seat are CON 28%(-20), LAB 35%(+4), LDEM 5%(-10), UKIP 30%(+24). Thanet, of course, was one of the areas where UKIP did particularly well in the local elections and is seen as a seat where Nigel Farage might stand at the next election. Note that there are some methodological changes from Survation’s past constituency polls. Previously they’ve weighted constituency polls by 2010 past vote and reallocated don’t knows based on past vote, in the same way they do for their national polls (though for practical reasons they do national polls online, but local polls by phone). For the latest polls they’ve changed method – no longer using political weighting, and not reallocating don’t knows. This is apparently part of a general review of how they do constituency polling, rather than something for this poll in particular.

Regular readers will be familiar with the debate over past vote weighting. Most companies (the primary exceptions being MORI and Opinium) weight their samples by both demographics, and by a political variable, normally how people voted at the last election, to ensure the sample is properly politically representative. While straightforward in theory, in practice this is complicated by the fact that poll respondents are not always very good at actually recalling how they voted at the last election (a phenomenon known as “false recall”). Companies that weight by past vote like ICM and ComRes therefore use a formula to estimate the level of “false recall” and account for that in their weighting schemes. Other companies, like MORI, take the view that false recall is so difficult to estimate and so potentially volatile that it renders past vote as unsuitable for weighting and risks cancelling out genuine volatility amongst the electorate, and therefore reject it completely.

In the case of the Survation poll of Thanet South, of the respondents who said they voted in 2010, about 41% said they voted Conservative, 38% Labour, 10% Lib Dem and 11% UKIP – so a three percent Conservative majority, when actually Laura Sandys had a seventeen percent majority. It underlines both the potential risk from not using political weighting, and the difficult choices that companies that do use it face – some of that difference will be false recall, but I suspect much of it is a sample that too Labour. Dividing one from the other is the challenge.

Survation have published a poll of Derby North and South Derbyshire constituencies for the Mirror and Progressive Polling, which shows a boost for Labour in both seats. Full tabs are here. The polling methodology itself was largely along ICM lines (telephone survey, weighted by past vote, ICM style weightings on likelihood to vote, and refusals and don’t knows reallocated according to how they say they voted last time).

In Derby North, currently a Labour marginal, the poll found figures of CON 23%(-9), LAB 51%(+18), LDEM 11%(-16), BNP 10%(+6), Others 6%(+3).

In South Derbyshire, currently a relatively comfortable Conservative seat, the poll found figures of CON 32%(-14), LAB 51%(+15), LDEM 9%(-7), BNP 1%(-3), Others 12%(+9)

The precise shares of the vote aren’t that interesting, given they are based on 174 and 149 people respectively, and hence have very large margins of error, but the overall picture is of a big swing towards Labour. Labour’s performance isn’t radically different from national polls, but there is a sharp drop in Tory support, something national polls aren’t showing. It implies that the Conservatives are doing worse in Derby than elsewhere.

Now, a plausible explanation is that this is connected with the Bombardier decision… it is certainly possible for large industrial closures to have big effects on particular seats (an excellent example is Redcar at the last election, where one assumes the closure of the steelworks was a factor behind the massive swing).

However, it pays to remember that correlation does not equal causality. This could be a Bombardier effect, it could be that the Conservatives are doing worse in a lot of urban seats, or in Midlands seats, or something entirely different. We don’t know. The rest of the questions are of no use in determining salience of the issue – questions of the “How likely is X to affect your vote at the election” pattern are worse than useless (they give false prominence to an issue with no requirement to balance it against other issues, or any measure of how likely people actually are to change their vote).

Of course, in practice we are probably years away from an election (in the Redcar example I quoted above the steel plant was mothballed a matter of months before the election) and there are likely to be major boundary changes in the Derby area (I’d expect Mid-Derbyshire to get the chop, with consequential knock on effects) anyway, but an interesting local straw in the wind nonetheless.

UPDATE: Unconnected to this (well, no more connected to this than every other post I make), there’s a very good explanation of sample error from Ben Goldacre here.