After the six point lead from Ipsos MORI, we now have a 17 point lead from Angus Reid. The topline figures, as I mentioned in my brief post below, are CON 39%(+1), LAB 22%(-2), LDEM 21%(+1). Others are unchanged on 18%. The poll was conducted over the weekend, so it is the first proper post-Queens speech poll.

The changes from Angus Reid’s last poll are all quite minor, showing a small shift away from Labour but nothing to get excited about. It does, of course, contrast sharply with MORI’s poll, particularly in terms of the level of Labour support.

Angus Reid are just entering UK polling so we don’t have a long track record to judge them by, but looking at their methodology I would expect them to have a tendency to show higher levels of Conservative and Lib Dem support and lower Labour support. I mentioned in my posts on MORI about “false recall”, people’s tendency to inaccurately report how they actually voted in 2005, and that ICM, Populus and ComRes all factor this into their weighting targets. AngusReid do not, implying no false recall at all, and the effect of this is that they weight the Conservatives and Lib Dems slightly higher and Labour slightly lower (though before people get carried away, it is pretty minor). From their three UK polls so far, they also seem to have a tendency to report significantly higher levels of support for minor parties than any of the other pollsters (though ComRes showed a similar figure in their last poll) – I can see no obvious methodological reason to explain the difference.

The bottom line is that on the short track record and methodology details we have I’d expect Angus Reid to show Labour a bit lower than companies like ICM and YouGov, so this poll is pretty much in line with the average Conservative lead still being somewhere around 13 or 14 points.

UPDATE: Forgot to say, we are also due a ComRes poll for the Independent… but not tonight. Presumably we can expect it later in the week.

Just to illustrate a post I made a fortnight or so ago. At the time Jackie Ashley was assuring us that Labour’s private polling “suggests that Labour could return to the Commons with just 120 MPs or thereabouts, taking the party back to 1930s territory.” To reduce Labour to 120 seats the Conservative would have needed a lead of about 28 points.

Today though, James Macintyre assures us that Labour’s private polling shows they are now ahead.

So… Labour have made up a deficit of around 28 points? In a fortnight? Either Labour’s private polling figures are so comically volatile as to be worthless, or one or both of you have been sold a pup, haven’t you?

Once again, please ignore any journalist claiming to have seen parties’ private polling showing X, unless they quote actual figures. If they do quote actual figures, go and contact the pollster responsible and request the release of the tables under the BPC disclosure rules so you can see for yourself what they actually say.

UPDATE: James now says Labour’s private polling doesn’t show them ahead. It shows a trend that he or his source interpret as showing that Labour will be in front by next year. Since there’s still no actual figures or tables for you to draw your own conclusions from, you should still ignore it.

There is a new(ish) Ipsos MORI poll in the Observer. The topline figures, with changes from MORI’s previous poll in mid-October, are CON 37%(-6), LAB 31%(+5), LDEM 17%(-2).

The poll was actually conducted last weekend at the same time as ICM’s Guardian poll, prior to the Queen’s speech, David Curry’s expenses accusations and the recent floods (before anyone suggests any of them might have contributed to it!). In ICM’s case the lead shrunk, but the Tory vote actually stayed the same, and the effect was to bring ICM in line with the sort of figures all the other pollsters were showing. This poll is clearly very different – it shows a 6 point collapse in Conservative support and represents by far the lowest Conservative lead for almost a year.

As ever, I would urge extreme caution on any poll showing a large shift in voting intention, especially where there is no obvious reason for a large, short term movement. Until we get the tables we can’t dig around to see exactly what is behind the movement, and until we find another poll supporting this shift, I wouldn’t get too excited/paniced (depending on your point of view!)