There is a new ComRes poll in tomorrow’s Independent, with topline figures of CON 36%(-3), LAB 25%(+3), LDEM 19%(+1). Others are collectively on 20%.
The topline figures would appear to show a swing back towards Labour – the 11 point lead is still enough for the Conservatives to secure an overall majority, but is the lowest Conservative lead since the end of last month. However, it’s important to note that ComRes have made a substantial change to their methodology this month.
Regular readers will know that ComRes used to use a method of past vote weighting that was quite confusing, and which seemed to result in them weighting to different targets each month. That’s now gone, and they are now weighting recalled past vote to a target made up 75% of the last general election result, and 25% the average of ComRes’s last 12 polls. My expectation is that this should result in some more consistent, less volatile figures.
ComRes’s new past vote weighting should on paper be almost identical to ICM’s method. Note that this doesn’t mean ComRes’s methodology is entirely comparable to ICM’s – there are still important differences. ComRes use a “squeeze question” to coax intentions out of people who don’t give a voting intention, ICM don’t. Secondly, ICM then rellocate 50% of don’t knows to the party they voted for in 2005. ComRes reallocate don’t knows to the party they identify with (and, as far as I can tell, they re-allocate all of them).
Other questions in the poll included which party people trusted more to “decide where spending cuts should be made” – 31% said the Conservatives, 21% Labour and 14% the Lib Dems, so pretty much in line with voting intention.
UPDATE: Full tables are here.