A new ICM poll in tomorrow’s Sunday Telegraph suggests that 39% of Liberal Democrat voters would prefer to have Charles Kennedy back as their leader, while 47% would still like a new leader (though bearing in mind that this is a normal sized sample, the sub-sample of Lib Dem voters is so small that the margin of error on the question will be huge.) Overall 80% of respondents said that the Lib Dems’ recent travails don’t make them any less likely to vote Liberal Democrat.

The topline figures, with changes from ICM’s poll conducted last weekend, are CON 37%(nc), LAB 36%(nc), LDEM 18%(-1). Part of any potential damage from Mark Oaten’s resignation would have been reflected in the poll conducted last weekend, but overall this suggests only slight damage. Matthew d’Ancona’s column in the Sunday Telegraph says that “The Lib Dems will be disappointed with their showing in our ICM poll today” – I rather suspect he’s wrong. After a torrid month I think the Liberal Democrats will be rather happy with 18% in ICM’s poll – at least they’ll be happier with this than the 13% and 15% they have in YouGov and MORI’s latest polls. That said compared to ICM’s polls last month, prior to the Liberal Democrats’ recent troubles, they are still down 3 points.

So, why the huge difference in the Lib Dem score between ICM and YouGov? The polls were conducted almost simultaneously and both pollsters’ figures are very stable. It is probably partially due to ICM’s weighting, which is slightly more favourable to the Lib Dems than other pollsters, and it is possible (though we won’t know until the full tables go up on ICM’s website) that it’s also partially due to ICM’s topline adjustment. Part of the reason for the decline in the Liberal Democrat’s vote in YouGov’s poll was that a larghe chunk of respondents who told YouGov they voted Lib Dem back in 2005 are now saying they don’t know how they’d vote. ICM’s topline adjustment means they assume that 50% of don’t knows vote the same way they did last time, potentially increasing Lib Dem support in the poll.

UPDATE: ICM’s topline adjustment to take account of don’t knows did indeed help the Liberal Democrats, although it only increased their reported level of support by 1 point, so it certainly doesn’t explain much of the 5 point difference.


Comments are closed.